Infectious Disease

T cells cannot see free-floating antigen. They can only recognize a peptide clamped by MHC and displayed on the cell surface — this constraint is called MHC restriction, and though it sounds abstract, it is the starting point of all adaptive immunity. To lay this skeleton bare takes only one sentence: where the antigen comes from, which route it travels, which MHC presents it, and which T cell is shown it. Follow this chain, and you will find that everything downstream — immunodeficiency, autoimmunity, tumor escape — is simply some link in this same chain going wrong.

Think of MHC as a display window on the cell surface. Class I displays "what the cell synthesizes itself" — viral proteins, tumor proteins, and self proteins alike, because all of them originate in the cytoplasm. These proteins are chopped up by the cytoplasmic proteasome, then carried through the TAP transporter into the endoplasmic reticulum to be loaded onto class I; the peptide is short, only 8–10 amino acids, because the class I groove is closed at both ends and cannot accommodate anything longer. Who is it shown to? CD8 T cells — the job of the CTL is to patrol the display windows of every nucleated cell in the body, and the moment it sees "something that should not be there," it kills that cell.

Class II works the other way around, displaying "what has been swallowed from outside" — bacteria, toxins, foreign proteins are first engulfed by phagocytes or dendritic cells into an endosome, packed into a lysosome and chopped up, and then HLA-DM swaps out the invariant chain (CLIP) that had been squatting in the groove and loads the foreign peptide in its place; this groove is open at both ends, so the peptide is long, 13–17 amino acids or even more. Who is it shown to? CD4 T cells — once a helper T cell sees it, it goes on to direct B cells and macrophages to act.

There is a well-known mnemonic here: class I × 8 = 8 (class I pairs with CD8), class II × 4 = 8 (class II pairs with CD4) — both products equal 8, so remembering one lets you derive the other. But more important than memorizing the trick is understanding why: class I corresponds to endogenous antigen because every single cell must be monitored for whether it has "gone bad" (hijacked by a virus, transformed into a tumor), so nearly every nucleated cell expresses class I; class II corresponds to exogenous antigen because it must coordinate the immune response, so only professional APCs — dendritic cells, macrophages, B cells — express it. Neutrophils are the shock troops charging the front line, not the messengers relaying orders — they are not major APCs and do not express class II, a favorite trap answer on the exam.

One high-frequency trap in passing: the MHC class III region encodes complement C2/C4/Bf and TNF, and plays no part in antigen presentation — a favorite decoy the exam loves to list among the presenting molecules. There is also a special case called cross-presentation: dendritic cells can "jump the queue" and load exogenous antigen onto class I to show CD8 cells — the key mechanism by which antiviral and antitumor vaccines can induce a CTL response.

T cell activation has an elegant design called the two-signal model. Signal 1 is the TCR's specific recognition of the MHC-peptide complex, with CD4 or CD8 acting as a co-receptor to stabilize the binding; Signal 2 is B7 (CD80/86) on the APC engaging CD28 on the T cell — this is called co-stimulation. Why must there be two signals? Because signal 1 alone is not safe enough: if the TCR misfires and recognizes a self-antigen, activating on that recognition alone would spell disaster for the organism. So evolution added an insurance policy — the cell must simultaneously receive a "danger signal" (B7 is expressed by the APC only in the context of infection) before it will act. If it receives only signal 1 without signal 2, the T cell enters a state called anergy — thereafter it will not respond even when given antigen again — an important mechanism of peripheral tolerance. The exam loves to bury a trap here, asking "what results from a missing signal 2?" — the correct answer is anergy, not activation.

After receiving both signals, there is still Signal 3 — the cytokine milieu — which decides which helper subtype CD4 will differentiate into: abundant surrounding IL-12 drives Th1, which secretes IFN-γ to ignite macrophages against intracellular bacteria and viruses; abundant surrounding IL-4 drives Th2, which makes IgE and recruits eosinophils to handle parasites and allergy; surrounding IL-6 + TGF-β + IL-23 drives Th17, which uses IL-17/22 to recruit neutrophils to guard the mucosa against extracellular bacteria and Candida; TGF-β paired with IL-2 drives Treg (master transcription factor FoxP3), which maintains tolerance via IL-10 and TGF-β. This is exactly why patients with STAT3-mutant Hyper-IgE (Job) syndrome, whose Th17 axis fails, suffer recurrent cold abscesses from Candida and Staphylococcus aureus — with Th17 gone, neutrophils can no longer be recruited.

Once a CD8 CTL recognizes a peptide on class I, it kills by direct contact, wielding three weapons: perforin, which punches a hole; granzyme B, which climbs through that hole to activate caspases and trigger apoptosis; and a second route, FasL (on the CTL) — Fas (on the target cell), which also triggers apoptosis. A third factor often mistaken for a "weapon" is IFN-γ, but it is actually an auxiliary — CTLs and Th1 cells secrete it to activate macrophages and upregulate MHC I on target cells; it is not a direct killing mediator. Also easily confused is ADCC (antibody-dependent cellular cytotoxicity): NK cells, macrophages, and eosinophils kill IgG-coated target cells that they recognize through Fc receptors; but the primary route for clearing virus-infected cells is direct CD8 CTL killing, not ADCC.

One last piece of scaffolding: CD3 carries ITAM (activating), while ITIM sits on KIR/PD-1 (inhibitory; CTLA-4 lacks a classic ITIM and works by competing for and removing B7). Clinically, anti-PD-1 agents (nivolumab, pembrolizumab) and anti-CTLA-4 (ipilimumab) work by releasing these inhibitory signals, letting T cells open fire on the tumor once again — a preview of the tumor immunology covered in Part B.

T cells pass through two sieves in the thymus: positive selection in the cortex — cells that bind self-MHC "weakly" survive, and this step decides MHC restriction; those recognizing class I become CD8, those recognizing class II become CD4. Negative selection in the medulla — cells that bind self-antigen "strongly" are killed off, a process called clonal deletion, the core of central tolerance.

The key point: under the control of the AIRE gene, medullary epithelial cells express an array of "peripheral tissue antigens" (proteins that should only appear in organs such as the pancreas or thyroid are rehearsed once more in the thymus), so that self-reactive T cells are deleted early. AIRE mutation → APECED (autoimmune polyendocrinopathy syndrome).

Following this logic: MHC class I deficiency (TAP1/2 defect) → class I fails to load peptide → CD8 cells cannot complete positive selection in the thymus → CD8 is markedly reduced; MHC class II deficiency = bare lymphocyte syndrome type II (CIITA/RFX defect) → CD4 cells cannot undergo positive selection → CD4 is markedly reduced. "CD4 remains normal" is a common trap answer — whatever positive selection selects for is what you end up having.

Adaptive immunity takes days to respond; innate immunity is on the scene within minutes to hours, following the logic of recognition (PRR) → response → recruitment → clearance. Every immune cell carries the same set of PRRs, recognizing conserved PAMPs on pathogens and DAMPs from injured cells. The division of labor runs as follows: TLR4 on the membrane recognizes LPS, TLR3/7/8 in the endosome recognize viral RNA, TLR9 in the endosome recognizes CpG DNA; in the cytoplasm, RLRs (RIG-I/MDA5) recognize RNA and cGAS-STING recognizes DNA (R pairs with R; c pairs with double-stranded D); the NLRP3 inflammasome activates caspase-1 to cleave and release IL-1β/IL-18.

Type I interferon is an easily misunderstood player. It does not kill the virus directly; rather, it is secreted by infected cells and pDCs to act on neighboring, uninfected cells, activating PKR/OAS/Mx so those neighbors enter an "antiviral state," suppress protein synthesis, and upregulate MHC I. IFN-α is a cytokine, not a chemokine; it belongs to innate, not adaptive, immunity; and it is secreted by infected cells and pDCs, not by activated T cells — three points the exam frequently inverts.

Leukocyte recruitment proceeds in four steps: selectin-mediated rolling → chemokine-triggered integrin activation → firm adhesion via ICAM → transmigration. Navigation relies on chemokine receptors, not TLRs; LAD-1 = CD18 (β2-integrin) deficiency → absent pus, delayed umbilical cord separation, a paradoxically elevated white count, and recurrent infection.

Complement's three pathways all converge on C3: classical (IgM/IgG–C1q), lectin (MBL recognizing mannose), and alternative (spontaneous hydrolysis) — all three generate C3 convertase → C3a + C3b, which then forms C5 convertase → C5b → C6789 = the MAC punching its hole. Along the way, C3b acts as an opsonin (recognized by CR1/CR3), and C5a acts as a chemoattractant and anaphylatoxin; complement is synthesized mainly by the liver and stands ready as zymogens. Deficiency correlations: C1/C2/C4 → SLE-like disease; C3 → severe pyogenic infection; C5-9 → recurrent Neisseria; C1-INH → HAE (elevated bradykinin); CD55/CD59 → PNH (CD59 blocks C9/MAC assembly).

NK cells belong to innate immunity but plug a gap left by adaptive immunity: the activating receptor NKG2D recognizes stress ligands, and the inhibitory receptor KIR recognizes MHC I — whether the cell kills or not depends on the balance between the two. When a virus or tumor downregulates MHC I to hide from the CTL, it is instead caught by the NK cell — this is missing-self. NK cells also carry out ADCC (CD16 recognizing IgG). Intestinal M cells merely sample and transport antigen to the underlying lymphoid tissue; they themselves neither kill microbes nor produce antibody (a point often quietly swapped out in exam questions).

A B cell's life splits into two stages: antigen-independent diversification (in the marrow) and antigen-driven optimization (in the germinal center). Diversity in the marrow is generated by V(D)J recombination, carried out by RAG1/2 (which recognizes the RSS and cuts the DNA) and TdT (which, after the hairpin at the break is opened, randomly adds N-nucleotides at the 3′ end, supplying "junctional diversity" — not "marking the break site," a common trap). Once heavy-chain rearrangement succeeds, it pairs with the surrogate light chain (VpreB + λ5) + Igα/β to form the pre-BCR; signaling depends on BTK — so a BTK mutation produces XLA, with B cells arrested at the pre-B stage. Pre-BCR signaling also triggers allelic exclusion: a single B cell expresses only one heavy-chain specificity.

In the germinal center (lymph node), driven by antigen and with T-cell help, AID steps in to do two things: somatic hypermutation (SHM), which fine-tunes affinity — acting only on the variable region, never the constant region (stating that it "also alters the constant region" is wrong); and class switch recombination (CSR), which swaps the Fc portion — changing function only, without adding antigen-binding diversity (stating that it "increases diversity" is also wrong). Diversity is chiefly decided back at V(D)J.

Structure decides function across the five antibody classes: IgM is a pentamer with a J chain — the first to appear in the primary response and the strongest activator of complement — but does not cross the placenta; IgG is a monomer, the only class that crosses the placenta, the mainstay of the secondary response, opsonization, and ADCC; mucosal IgA is a dimer plus a J chain (made by the plasma cell) plus a secretory component (derived from epithelial pIgR) — the secretory component is contributed by the epithelium, not the plasma cell; IgA class switching depends on IL-10/TGF-β (not IL-12); IgE binds FcεRI on mast cells and drives allergy and antiparasitic responses; IgD together with IgM forms the BCR on the surface of mature naive B cells. Different IgG subclasses have different affinities for the same Fc receptor (IgG3 > 1 > 4 > 2 for FcγRI) — the claim that "subclasses are equal at the Fc receptor" is wrong.

The final contrast is T-dependent versus T-independent antigen: proteins are TD, requiring T-cell help, producing a germinal center, and yielding SHM/CSR/memory; polysaccharides (such as capsules) are TI — no T-cell help needed, no memory, dominated by IgM — which is exactly why children under 2 respond poorly to plain polysaccharide vaccines and need a conjugate that attaches the polysaccharide to a protein carrier, converting it into a TD antigen. This chain leads directly into the vaccine story of Act Two.

Now that the skeleton of recognition from Act One is standing, Act Two does two things: first, it lets you work backward from what kind of infection recurs to which arm of immunity is missing (the entire face of primary immunodeficiency); second, it turns that same logic around and applies it to vaccine design — a vaccine is simply "using a safe antigen to train one specific arm." We then close with three major clinical applications: transplantation, tumor immunity, and self-tolerance.

There are hundreds of primary immunodeficiencies, but the exam's core logic reduces to one sentence: the pattern of infection is the fingerprint of the immunodeficiency. Seize on "the recurring pathogen plus the age of onset," and half the question is already solved.

SCID is a severe deficiency of T cells, often taking B cells or NK cells down with it. The most common form is X-linked SCID — an IL-2Rγc mutation (this γ chain is a shared receptor component for IL-2, 4, 7, 9, 15, and 21), clinically T⁻B⁺NK⁻; another is ADA deficiency, in which toxic dATP wipes out lymphocytes wholesale, T⁻B⁻NK⁻. SCID is an absolute contraindication to live vaccines (BCG and rotavirus vaccine can be fatal), and requires bone marrow transplantation.

Recall the earlier comparison: MHC class II deficiency (BLS-II, CIITA/RFX) → CD4↓, class I/TAP deficiency → CD8↓; ataxia-telangiectasia (AT) = ATM mutation → both V(D)J recombination and DNA repair fail → B/T↓, AFP↑, radiosensitivity, and increased malignancy risk.

Four antibody-centered contrasts: XLA = BTK (B cells stuck at pre-B, all Ig classes low); CVID (adult onset, B cells cannot mature into plasma cells); selective IgA deficiency (the most common PID, mostly asymptomatic, but transfusion can trigger anaphylaxis); Hyper-IgM = CD40L/CD40/AID (no class switching, so IgM↑, IgG/A/E↓; the CD40L form also develops PCP).

Gene-syndrome pairings for T cell signaling are a classic exam point: DiGeorge = 22q11.2 deletion (third/fourth pharyngeal pouch → absent thymus/parathyroids → hypocalcemic seizures + conotruncal cardiac defects + facial anomalies + T↓); Wiskott-Aldrich = WASp (cytoskeleton, not NF-κB) → eczema + thrombocytopenia with small platelets + recurrent infection; Hyper-IgE/Job syndrome = STAT3 → Th17↓ → cold abscesses, Candida; MSMD = IFN-γR → macrophages cannot be ignited → disseminated mycobacteria/Salmonella; NEMO = the NF-κB axis. Writing WAS as NF-κB is a classic point-losing error.

Phagocytes: CGD = NADPH oxidase deficiency (no respiratory burst) → recurrent infection with catalase-positive organisms; confirmed by a negative DHR/NBT test. Chediak-Higashi syndrome = LYST (giant granules, albinism, neuropathy); LAD-1 = CD18, covered earlier.

Finally, never forget HIV, the great villain of secondary immunodeficiency: gp120 binds CD4 plus a co-receptor (CCR5/CXCR4), driving a progressive decline in CD4. Patients in the chronic phase do generate CD8 CTLs and anti-HIV antibodies — the problem is that their function fails and the virus escapes — so "no response is generated at all" is wrong. CD4 < 200 marks the onset of AIDS, bringing opportunistic infections such as PCP, toxoplasma encephalitis, and MAC.

If PID is "one arm broken," a vaccine is the reverse — using a safe antigen to train that specific arm. The central tension is immunogenicity (must be strong) versus safety (must be stable). Live vaccines are potent but carry a risk of reversion to virulence; inactivated/subunit vaccines are safe but often require adjuvants and boosters.

The single highest-yield concept is the mechanism of the conjugate vaccine, which picks up directly on the TI-vs-TD groundwork laid at the end of Act One: a pure capsular polysaccharide is a TI antigen → infant B cells have no memory for it → attach it to a protein carrier → it becomes TD → help arrives via Tfh + CD40-CD40L → memory and class switching (high-titer IgG) result. What the T cell recognizes is a peptide from the protein carrier, not the polysaccharide itself — a detail frequently tested. Hence plain polysaccharide vaccines (PPSV23) are effective only from age 2 onward and produce no booster effect; conjugate vaccines (PCV13/15/20, Hib, MCV) can be used in infants and young children.

The route of administration also determines the type of immunity produced: oral/mucosal → sIgA (rotavirus, OPV); intramuscular/subcutaneous → systemic IgG (most injected vaccines). Giving only an intramuscular vaccine against an enteric pathogen prevents paralysis but leaves intestinal sIgA insufficient, unable to fully block replication and shedding — which is exactly why the polio eradication campaign once relied heavily on OPV (despite its VAPP risk).

Other key clinical points: pertussis now uses aP (acellular), replacing the old wP; the only natural host of poliovirus is humans (it does not infect cattle), which is the key fact behind its eradicability; HPV-16/18 account for ≈70% of cervical cancer, and the vaccine prevents virus-associated cancer; measles has a high R0 (12–18) and needs roughly 95% vaccination coverage to reach herd immunity — and "herd immunity" is indirect protection at the population level, not individual natural immunity; don't conflate the two.

Transplant rejection can be diagnosed instantly by its timeline: hyperacute (minutes to hours) = preformed antibody + complement (seen right away with ABO/HLA mismatch); acute (days to weeks) = T-cell mediated (responsive to immunosuppressants); chronic (months to years) = vascular intimal thickening and fibrosis (poor response to treatment).

GVHD (graft-versus-host disease) runs in the opposite direction: donor T cells attack the recipient (skin, liver, gut), commonly seen after bone marrow/hematopoietic stem cell transplantation. Prevention relies on depleting mature donor T cells (anti-CD52 alemtuzumab, ATG) and HLA matching. The MLR (mixed lymphocyte reaction) does not reduce GVHD — it is an in vitro tool for measuring the strength of a T-cell response, and it actually activates T cells; a frequently tested trap.

The core of tumor immunology is the three phases of immunoediting: elimination → equilibrium → escape (the immune system "prunes" the tumor, it does not "shrink" it). Tumor antigens fall into two classes: TSAs (mutant neoantigens, HPV E6/E7) and TAAs (overexpressed normal genes, such as tyrosinase in melanoma, MAGE, CEA, AFP, HER2/neu). Tumor escape works by downregulating MHC I (evading the CTL but getting caught by NK cells instead), expressing PD-L1, secreting TGF-β/IL-10, and recruiting Treg/MDSCs.

Matching immunotherapy drugs to their targets is an easy-point question: trastuzumab → HER2/neu (breast cancer; blocks signaling + ADCC); rituximab → CD20; ipilimumab → CTLA-4; nivolumab/pembrolizumab → PD-1; CAR-T (engineered T cells, such as anti-CD19 for hematologic malignancies). One critical point: CTLA-4 sits on the T cell (not the tumor cell — the exam's favorite way to write it wrong); it competes with CD28 for the APC's B7 and wins, thereby suppressing activation; PD-L1, by contrast, is the ligand commonly found on the tumor cell.

Last comes self-tolerance. Central tolerance uses negative selection in the thymus (for T cells) and marrow (for B cells) to eliminate self-reactive cells; the mainstay of peripheral tolerance is the Treg — surface markers CD4⁺CD25⁺, master transcription factor FoxP3 — acting through three mechanisms: ① secreting IL-10/TGF-β/IL-35; ② CTLA-4 competing for B7 (contact-dependent inhibition); ③ consuming IL-2 to drive effector T cells into apoptosis. The Treg does not rely on cytotoxic killing — that is the CTL's job. FoxP3 mutation → IPEX (multi-organ autoimmunity); IL-10 deficiency → IBD; CTLA-4/FasL deficiency → autoimmunity.

Oral tolerance is a special case of peripheral tolerance: a high dose → T-cell anergy; a low dose → induces Tregs that secrete TGF-β (not because "the molecule is too small," and not because "it is presented in the thymus").

Breaking tolerance operates through four mechanisms: genetic susceptibility (HLA associations, such as B27→ankylosing spondylitis, DR3/4→T1DM/SLE, DR4→RA), molecular mimicry (streptococcal M protein → rheumatic fever), exposure of a sequestered antigen (sympathetic ophthalmia), and transplacental antibody transfer (maternal IgG autoantibodies → transient neonatal autoimmunity, as in neonatal Graves' disease or neonatal SLE). One interesting contrast: T1DM is mainly cell-mediated (CTLs attacking β cells) and does not rely on IgG crossing the placenta to reach the fetus, so a mother with T1DM does not "transmit" diabetes to her newborn.

The SLE pathway of anti-dsDNA → TLR-9 → IFN-α is a high-frequency mechanism question: apoptotic cells release dsDNA → it forms immune complexes with antibody → these engage TLR-9 inside the endosomes of B cells/pDCs → triggering type I interferon → which amplifies the autoimmunity. CD23 is the low-affinity IgE receptor and has nothing to do with autoimmune tolerance — don't be fooled when it is listed among the options.

Act Three brings us down to the clinical terrain. When immune recognition misjudges "harmless" as "mortal enemy," the result is allergy; when it misjudges "streptococcal M protein" as "heart valve," the result is rheumatic fever; when recognition itself is sound but the pathogen is simply too cunning, the result is the game of hide-and-seek that is gynecologic infection and parasitism. Once we walk through these stories, Part A comes to a close.

For any immune-mediated disease, ask three things first: who is acting (antibody or cell) + against what (the antigen) + by what mechanism of injury. The Gell-Coombs four-type framework:

Only Type IV does not depend on antibody, so it cannot be passively transferred by serum. Type II further splits into two subtypes: ① cytotoxic (complement/ADCC, as in hemolysis or Goodpasture syndrome); ② receptor-mediated (the antibody does not kill the cell — it only alters receptor function) — Graves' disease (anti-TSHR stimulation → hyperthyroidism) and myasthenia gravis (anti-AChR blockade → weakness) both belong to Type II, not Type I or III, and the exam often flips this.

The details of Type I are the most frequently tested — remember that the full causal chain runs five steps, not just two.

The Th2 cytokine network is the engine of allergy: IL-4 (IgE), IL-5 (eosinophils), IL-13 (mucus + airway hyperreactivity); IFN-γ belongs to Th1 and is not a feature of allergy.

The hygiene hypothesis: an overly sanitized modern environment → infants and young children lack microbial stimulation → insufficient Th1, a skew toward Th2 → rising allergy rates. So the statement "prolonged antibiotic use reduces asthma" runs opposite to the hygiene hypothesis and is false. The house dust mite is the single most important inhalant allergen; its allergen comes from its excreta and body proteins, it favors warm, humid conditions, and it is visible under an ordinary light microscope (an electron microscope is not needed).

The first line for anaphylaxis is epinephrine 0.3–0.5 mg IM (1:1000, injected into the anterolateral thigh, the vastus lateralis), repeatable every 5–15 minutes. Its mechanism is α1-mediated pressor effect, β1-mediated inotropy, and β2-mediated bronchodilation plus suppression of mast cell release — one drug reversing both hypotension and bronchoconstriction at once. Neither antihistamines nor corticosteroids are life-saving drugs — they only address skin itching and the delayed reaction; the only agent that reverses hypotension and laryngeal edema is IM epinephrine (not IV, not subcutaneous, unless cardiac arrest has already occurred, in which case IV is used).

Chronic spontaneous urticaria and HAE are an easily confused pair: urticaria works through histamine, itches, and responds to antihistamines (first line is a second-generation, non-sedating antihistamine such as cetirizine, loratadine, or desloratadine; if ineffective, quadruple the dose; if still ineffective, add omalizumab, an anti-IgE agent); HAE works through bradykinin, does not itch, and does not respond to antihistamines — as noted earlier, C1-INH deficiency sends bradykinin soaring, and treatment requires a C1-INH concentrate or icatibant (a bradykinin antagonist). Treating HAE as ordinary urticaria with antihistamines is a classic pitfall.

The pathogenic chain: group A streptococcal (GAS) pharyngitis → 2–4 weeks after infection → anti-streptococcal antibodies cross-react (molecular mimicry, Type II) with myocardial/valvular antigen → inflammation spreads through connective tissue body-wide. The major Jones criteria are remembered by JONES ♥: Joint (migratory polyarthritis, the most common), ♥ (O) carditis (the most severe, the only one leaving a lasting valvular sequela, especially of the mitral valve), Nodules (subcutaneous), Erythema marginatum (<5% but highly specific), and Sydenham chorea (late-onset, may occur in isolation). Minor criteria: fever, arthralgia, elevated ESR/CRP, prolonged PR interval. Diagnosis = evidence of antecedent GAS infection + 2 major criteria, or 1 major + 2 minor; acute-phase treatment is penicillin plus an NSAID, followed by long-term penicillin prophylaxis against recurrence (because reinfection worsens the valve).

Numeric traps: it is group A (not B), the interval is 2–4 weeks (not 3 months), and among those not properly treated, 30–50% progress to rheumatic heart disease (not 10%).

A quick differential for lower genital tract infection:

Both BV and trichomoniasis are treated with metronidazole, but only trichomoniasis is an STI requiring the partner to be treated as well. PID is an ascending infection, most often caused by Neisseria gonorrhoeae plus Chlamydia, presenting with cervical motion tenderness + adnexal tenderness + fever; its complications chain onward to salpingitis → infertility/ectopic pregnancy/chronic pelvic pain/Fitz-Hugh-Curtis syndrome (perihepatitis with violin-string adhesions)/tubo-ovarian abscess; treatment uses broad coverage (ceftriaxone plus doxycycline, ± metronidazole).

The management of asymptomatic bacteriuria (ASB) in pregnancy is a high-frequency easy-point question: urine culture ≥ 10⁵ cfu/mL is usually left untreated in non-pregnant women, but in pregnant women it must always be treated and followed up — because pregnancy dilates the ureters, compresses the bladder, and causes urinary stasis, 30–40% progress to acute pyelonephritis, which can cause preterm birth and low birth weight. Pregnancy turns a harmless bacterium into the prelude to pyelonephritis.

The most efficient way to learn parasitology is the single chain "organism → life cycle/transmission → site of infestation → characteristic lesion → diagnosis." First, malaria:

Cytoadherence in malignant malaria: infected red cells express PfEMP1, which adheres to the endothelium via ICAM-1, CD36, and thrombospondin → capillary obstruction → cerebral malaria. Blackwater fever is massive intravascular hemolysis involving even uninfected red cells, with hemoglobinuria turning the urine black. Treatment: uncomplicated disease is treated first-line with an ACT (artemisinin-based combination therapy); vivax/ovale require adding primaquine to clear the hepatic hypnozoite (check G6PD before use); severe disease or cerebral malaria is treated first-line with IV artesunate (superior to the older quinidine). In one sentence: malignant malaria kills through endothelial adherence; vivax/ovale relapse from sleeping in the liver.

The exam point for amoebae is the nuclear count of the intestinal cyst: E. histolytica (pathogenic) = 4 nuclei, Entamoeba coli (non-pathogenic, the largest) = 8 nuclei, E. hartmanni (non-pathogenic, the smallest); Acanthamoeba keratitis = contact lenses + tap water/pool water, with corneal scraping showing a double-walled, angular cyst. E. histolytica produces flask-shaped ulcers and a right-lobe liver abscess (anchovy-paste pus); treatment is metronidazole plus paromomycin to clear luminal cysts.

Other high-frequency protozoa: Toxoplasma (definitive host is the cat; primary infection during pregnancy causes congenital infection with intracranial calcification, chorioretinitis, and hydrocephalus; in AIDS with CD4<100 it causes ring-enhancing encephalitis); Cryptosporidium (severe chronic watery diarrhea in AIDS; nitazoxanide for immunocompetent patients, ART-driven immune recovery in AIDS; traditionally no specific drug, round oocysts positive on acid-fast stain); Leishmania donovani (visceral leishmaniasis, or kala-azar, invading liver, spleen, and marrow; transmitted by the sandfly); Giardia (steatorrhea, duodenal, treated with metronidazole).

The backbone for trematodes (flukes) is their two transmission routes: schistosome cercariae penetrate the skin directly (no metacercarial stage), whereas other flukes depend on being eaten raw (as encysted metacercariae); the first intermediate host for every fluke is a snail. Three schistosome species: S. japonicum (superior mesenteric veins, highest egg output, >3000/day, the most severe Katayama fever, eggs can reach the brain); S. mansoni (inferior mesenteric veins, ~300/day); S. haematobium (the vesical venous plexus → chronic infection causing squamous cell carcinoma of the bladder). Katayama fever is acute-phase immune-complex deposition (a Type III, serum-sickness-like reaction).

A pairing mnemonic for food-borne flukes: fish → the liver fluke (Clonorchis, cholangiocarcinoma); crab/crayfish → the lung fluke (can migrate ectopically to the brain, causing hemoptysis mimicking tuberculosis); aquatic plants → the giant intestinal fluke (stays confined to the small intestine, never reaches the brain, the least likely to migrate ectopically) / Fasciola (the sheep liver fluke, from eating raw watercress, causing halzoun syndrome — the adult fluke attaches to the throat). Diagnosis relies on finding eggs in the stool (operculated, except for schistosome eggs, which lack an operculum); S. haematobium is found via eggs in the urine; the tissue-invasive stage commonly shows peripheral eosinophilia plus elevated IgE. Praziquantel is first-line for most flukes; the exception is Fasciola hepatica, treated with triclabendazole.

Exam questions on parasites and their vectors are terrifying because the names sound alike, the host relationships sound alike, and the transmission vehicles get more tongue-twisting one after another. But return to a single shared reasoning chain — "does the human ingest an egg or a larva, is the human the definitive host or the intermediate host, where does the worm live, and what drug is used" — and every seemingly scattered detail falls into line on its own. What matters for a parasite is not its name but its route; what matters for an arthropod is not its taxonomy but whether it "flies or crawls." Work through these two routes thoroughly, and this chapter is won.

To tell tapeworms apart, first ask the simplest question: does the human ingest an egg or a larva (a cysticercus hiding in meat)? Eating the larva, the worm matures into an adult in the gut, the human is the definitive host, and the result is intestinal taeniasis. Eating the egg, the hexacanth larva burrows through the intestinal wall and travels via the bloodstream into tissue to form a cyst, and the human is treated as an intermediate host, growing a cyst inside an organ. Get this one principle and nearly the whole chapter falls into place.

The beef tapeworm has only the meat route, so the human is always the definitive host and symptoms stay confined to the gut; its scolex is unarmed, bearing suckers only, and its gravid proglottid has a highly branched uterus. The pork tapeworm is the sole exception to this principle — both routes work for it: eat raw pork containing the cysticercus, and the worm matures into an adult in the gut, with the human as definitive host; but accidentally swallow the pork tapeworm's eggs instead (say, from food contaminated by a carrier's feces), and the hexacanth larva burrows through the gut into the blood and travels to the brain to form a cyst, making the human an intermediate host with neurocysticercosis — one of the world's major infectious causes of epilepsy. So for the superintendent in the opening scene, those small calcified cysts in his brain were not caused by eating meat — they were caused by swallowing eggs, and that is exactly the trap the exam loves to dig.

The Nordic engineer belongs to a different storyline. The fish tapeworm (Diphyllobothrium latum) lodges in the proximal jejunum and competes with its host for vitamin B12 — the fatter the worm grows, the scarcer the B12 becomes, hence the megaloblastic anemia. This is exactly why a serving of smoked raw fish can breed a systemic blood disorder. It can grow beyond ten meters, making it the longest tapeworm found in humans, seen mainly in the raw-freshwater-fish cultures of Scandinavia and Japan.

Echinococcus is another organism the exam loves to trap you with. Its definitive host is the dog, and humans become infected only by accidentally ingesting something contaminated with dog feces — so what the human swallows is the egg, not meat. The hexacanth larva travels via the bloodstream to the liver (most often) or the lungs, forming a hydatid cyst that is beautiful but lethal. The single most important rule in its management: never attempt a rash needle aspiration, because once the cyst fluid leaks, it is equivalent to scattering a swarm of daughter cysts through the peritoneal cavity, and it can also trigger anaphylactic shock. Standard management is careful surgical removal capped off with albendazole. Neurocysticercosis is likewise treated with albendazole, combined with corticosteroids to prevent cerebral edema.

As for the dwarf tapeworm (Hymenolepis nana), its most distinctive feature is that it is the only tapeworm needing no intermediate host at all — once its egg is swallowed, it can complete its entire life cycle within the small intestinal villi, and it can even autoinfect. The dog flea tapeworm (Dipylidium caninum) is another example where the intermediate host is easily overlooked: humans fall ill only by swallowing a flea carrying the cysticercoid, most often young children who play often with cats and dogs. For treatment, praziquantel is the choice for most intestinal tapeworms, while albendazole is chosen for Echinococcus and neurocysticercosis.

The clinical picture of a nematode infection is decided by the worm's route of migration. Whether entering by mouth or by burrowing through skin, the larvae of both roundworms and hookworms migrate through the lungs, so early infection may show pulmonary infiltrates plus eosinophilia (Löffler syndrome); whipworm never passes through the lungs and settles only in the large intestine, so heavy infection can produce rectal prolapse; pinworm crawls out of the anus at night to lay its eggs, so it presents as nocturnal anal itching, diagnosed by the morning perianal tape test, while stool microscopy actually misses it — secure these several basic easy-point questions first.

Strongyloides stercoralis is the most dangerous organism in this section, because it can autoinfect within the intestine itself. The filariform larva burrows through the skin, reaches the gut, matures into an adult, and the larvae it produces can mature again inside the gut and burrow back through it into the blood — forming an internal loop that lets a person carry the worm asymptomatically for decades. Once immunity is suppressed (corticosteroids, HTLV-1, HIV), this loop spins out of control, the worm burden explodes, and larvae drag intestinal bacteria along with them into the bloodstream, causing multi-organ dissemination plus gram-negative sepsis. The deadliest trap lies here: during dissemination, eosinophils paradoxically fail to rise significantly, because the host's immunity has already been crushed, so "no eosinophilia" can never be used to rule out a parasitic infection. Ivermectin is first-line treatment.

Angiostrongylus cantonensis is another classic piece of reasoning. A human swallows an L3 larva carried by a snail, a slug, or raw greens they crawled over, and the larva migrates to the central nervous system, causing eosinophilic meningitis — abundant eosinophils in the cerebrospinal fluid should bring this organism to mind. But this worm cannot complete its life cycle in a human, because the human is an accidental host; the true definitive host is the rat — so inside a human the worm simply takes a wrong turn and gets lost in the central nervous system.

Trichinella's story is "the adult in the gut, the larva in the muscle." A human eats raw pork containing the encysted larva; the adult rapidly produces a large brood of larvae in the small intestine and then disappears, while the larvae travel via the bloodstream throughout the body to encyst in striated muscle, producing the signature trio of myalgia, fever, and periorbital edema plus marked eosinophilia. Remember the diagnostic trap: looking for eggs in the stool is useless, because the intestinal stage is brief and produces no eggs; confirmation relies on a muscle biopsy to find the encysted larva, or on serology.

Onchocerca volvulus moves the stage to the riverbank. The blackfly bites a human, and the microfilariae migrate to the subcutaneous tissue and the eye, producing subcutaneous nodules and hanging groin, with the most lethal outcome being keratitis and iritis that ultimately end in blindness — this is "river blindness." Diagnosis relies on a skin snip to find the microfilariae. The exam loves to dig one particular trap: a question stating "Onchocerca causes pulmonary nodules" — wrong, it does not cause pulmonary nodules. Treatment is ivermectin, but when it coexists with a high microfilarial burden of Loa loa, watch out for the risk of fatal encephalopathy. Both Gnathostoma spinigerum and Capillaria philippinensis are linked to raw freshwater fish: the latter is ingested as a larva (not an egg) and causes protein-losing enteropathy, while the former causes migratory subcutaneous and periorbital swellings.

The tsetse fly and the kissing bug are the pair most loved for a swap in this section. The tsetse fly lives in Africa, and its bite injects T. brucei directly into the bloodstream, causing African trypanosomiasis that ends in lethargy — hence "sleeping sickness." The kissing bug lives in the Americas, and it does not inject the organism through the bite itself — instead it defecates while feeding, and when the person scratches the bite, the feces get rubbed into the broken skin or a mucous membrane, and T. cruzi enters the body through that route, causing Chagas disease. Remember one line: "tsetse bites in Africa, the kissing bug defecates in the Americas" — the route of transmission differs and the continent differs, so a question that writes the tsetse fly as transmitting Chagas disease is simply wrong.

Both ticks and lice can transmit relapsing fever: the soft tick (Ornithodoros) transmits endemic relapsing fever, while the body louse transmits epidemic relapsing fever, and the causative organism in both is a Borrelia spirochete. Ixodes, among the hard ticks, is "the king of one-vector, many-diseases" — the very same deer tick can transmit Babesia, Lyme disease, and Anaplasma all at once, so co-infection is far from rare. The chigger mite hides a detail the exam is extremely fond of testing: only the larval stage actively bites humans to feed on tissue fluid; the nymph and adult stages feed on plants and small insects, so scrub typhus is transmitted only during the larval stage; the eschar left at the bite site, together with fever and lymphadenopathy, is its signature.

As for the division of labor among the three mosquito genera, this is an easy-point pairing on vector questions, and getting it backward costs you the point outright. Anopheles mosquitoes transmit malaria (Plasmodium); Aedes mosquitoes (day-biting, breeding in standing water in containers) transmit dengue, yellow fever, Zika, and chikungunya; Culex mosquitoes transmit Japanese encephalitis and Wuchereria bancrofti. The blackfly transmits river blindness, and the sandfly transmits visceral leishmaniasis (kala-azar) — these sit on the same axis as the nematode and protozoan sections earlier, since a worm's journey begins right in the vector's mouth. One last easily confused set of names: "kala-azar" is visceral leishmaniasis transmitted by the sandfly, "blackwater fever" is a complication of malignant malaria, and "the Black Death" is the plague transmitted by the rat flea — all three carry the word "black" and are all deadly, yet they are entirely different diseases.

Exam questions on autoimmune disease are difficult because a single tube of blood can turn up a dozen or more autoantibodies, each with its own matching disease. But return first to one principle — "direction, specificity, mechanism" — and every seemingly scattered antibody falls into place on its own. Direction means that during inflammation, C3 and C4 fall because they are being consumed, not because they are rising; blood cells fall in number because they are being attacked by autoantibodies, not because they are increasing. Specificity means that a screening antibody (such as ANA) is sensitive but not specific, and pinning down the diagnosis requires a more specific antibody. Mechanism means matching the drug to whichever cell-signaling pathway or cytokine is actually involved, rather than "reaching for a biologic the moment you see a rheumatic disease."

Translated into the blood count, SLE is a story of "all three lineages falling." Red cells are attacked by anti-red-cell antibody, producing hemolytic anemia with a positive Coombs test; white cells are suppressed by autoantibody and marrow suppression; platelets are destroyed by antiplatelet antibody — so the pattern is anemia, leukopenia, and thrombocytopenia. A question stating "leukocytosis" or "increased platelets" is always wrong. Add to this that C3 and C4 fall rather than rise during the active phase, and the directionality question is won.

Choosing among the autoantibodies is where it is easiest to get lost. ANA is a screening tool — extremely sensitive, with a negative result nearly ruling out SLE, but not highly specific: low-titer positives are common in healthy people (about 20–30% at 1:40, about 5% at 1:160), so "ANA has 90% specificity" is a numeric trap. Pinning down the diagnosis relies on more specific antibodies: anti-dsDNA has >95% specificity and correlates best with lupus nephritis and disease activity, so it can be used to track the course of disease; anti-Sm has the highest specificity but low sensitivity, does not track with activity, and is a lifelong marker. Other antibodies each have their own specialty: anti-Ro/La crosses the placenta to cause neonatal lupus and congenital heart block; anti-histone appears in drug-induced lupus (hydralazine, procainamide, INH, and others), mostly resolving after the drug is stopped, with renal and CNS involvement rare; antiphospholipid antibody is linked to thrombosis and miscarriage, yet paradoxically prolongs the APTT in vitro.

The core definition of JIA is simple: onset before age 16, arthritis of one or more joints persisting ≥6 weeks, with other causes excluded. But subtyping looks at the joint count within the first six months — and this is exactly the trap the exam loves to dig: "6 weeks" is the duration threshold for JIA arthritis as a whole, while "6 months" is the counting window for subtyping, and the two "sixes" cannot be swapped. Four or fewer joints within the first six months is oligoarthritis; five or more is polyarthritis — this window decides the subtype, not the diagnosis.

Each subtype has its own signature: oligoarthritis is often ANA-positive and favors young girls, and the complication most worth remembering is chronic, "asymptomatic" anterior uveitis — which is why regular slit-lamp follow-up is required, rather than waiting for the patient to complain of a red eye. Enthesitis-related arthritis is associated with HLA-B27 and favors boys, and runs exactly the opposite way, causing acute, "symptomatic" anterior uveitis, with redness, pain, and photophobia that the patient will report on their own. One quiet, one loud; one negative, one positive — this contrasting pair is the single most frequently tested point. The signature of systemic-onset JIA (Still disease) is a daily spiking fever plus a salmon-colored rash, with enlargement of the liver, spleen, and lymph nodes, and the complication most to be feared is macrophage activation syndrome (MAS). RF-positive polyarthritis resembles adult RA and carries a poor prognosis; psoriatic and undifferentiated arthritis are the remaining two subtypes, bringing the total to seven.

The choice of biologic follows directly from the mechanism. Polyarticular disease is treated with an anti-TNF-α agent (etanercept, adalimumab) or abatacept (CTLA4-Ig, which blocks B7-CD28 co-stimulation); systemic-onset disease favors an anti-IL-6R agent (tocilizumab) or an anti-IL-1 agent (canakinumab, anakinra), because the core signal in Still disease lies squarely in the IL-1/IL-6 axis. The trap option is omalizumab — it is an anti-IgE agent used for asthma and urticaria, entirely unrelated to JIA, and should be crossed out on sight.

The five major manifestations of the Jones criteria are captured by the five letters of JONES: J is Joints, migratory polyarthritis affecting the large joints (knee, ankle, elbow, wrist) — cross out immediately any question stating "mainly affects the small joints"; O is carditis, pancarditis, with the mitral valve most often affected and also the most severe long-term sequela; N is subcutaneous Nodules; E is Erythema marginatum; and S is Sydenham chorea. Diagnosis requires two major criteria, or one major plus two minor, and there must be evidence of antecedent GAS infection (a rising ASO titer, a positive throat culture, and the like).

Remember the three features of the arthritis: migratory, involving large joints, and an excellent response to aspirin. For treatment, penicillin clears the GAS in the acute phase; secondary prophylaxis uses long-acting benzathine penicillin G, given IM every three to four weeks, continued into adulthood or even for life in those with carditis or valvular disease, because one more encounter with GAS will damage the valve even further.

The first key to pediatric rash questions is the timing relationship between the rash and the fever. "Fever first, then rash — the rash erupts as the fever breaks" describes roseola infantum (HHV-6), confirmed further by Nagayama spots at the junction of the uvula and soft palate; "rash appearing while the fever continues" describes measles or rubella. Measles spreads from behind the ears to the face and then the trunk, and Koplik spots on the buccal mucosa are a signature unique to measles, with the "3 C's" (cough, coryza, conjunctivitis) being classic; rubella resolves within two to three days, its signature being swollen postauricular and occipital lymph nodes, and the real danger lies in maternal infection during pregnancy, which causes congenital rubella syndrome (cataracts, cardiac defects, deafness).

Varicella (chickenpox) is "a sky full of stars" — papules, vesicles, and crusts all visible at the same time, distributed mainly on the trunk and spreading centrifugally to the limbs. The signature of erythema infectiosum (fifth disease, parvovirus B19) is a slapped-cheek rash, but the truly dangerous populations are pregnant women (fetal hydrops) and patients with sickle cell disease (aplastic crisis), because B19 infects red cell precursors. The ulcers of hand-foot-mouth disease/herpangina (Coxsackie A16, EV71) hide on the posterior pharyngeal wall — a sharp positional contrast with HSV gingivostomatitis, which sits at the front of the mouth (gums, tip of the tongue); don't mix up front and back. Remember measles complications on three levels: the most common acute-phase complication and cause of death is pneumonia; encephalitis can also complicate it; and years later, a rare but fatal outcome is SSPE (subacute sclerosing panencephalitis).

The shared story of the herpesviruses is latency and reactivation. After VZV resolves, it lies latent in the dorsal root and cranial nerve ganglia, and reactivating decades later along a single dermatome is shingles — a trick that belongs to the DNA herpesviruses, one that measles, rubella, and mumps (all RNA viruses) cannot pull off. EBV infects B lymphocytes via the CD21 receptor, causing infectious mononucleosis, and its most famous trap is that giving ampicillin/amoxicillin to an IM patient produces a body-wide maculopapular rash in nearly 100% of cases, yet this is not a true allergy. CMV is the one most often tested backward: roughly 90% of congenital CMV infections are asymptomatic, with only about 10% showing hepatosplenomegaly, jaundice, and petechiae, and the most common sequela being hearing loss; a question stating "90% are symptomatic" is wrong.

Kawasaki disease is a medium-vessel arteritis, and the complication most to be feared is a coronary artery aneurysm. Diagnosis requires fever for at least five days plus at least four of the five CRASH features: Conjunctivitis, bilateral non-purulent; Rash, polymorphous; Adenopathy, cervical, often unilateral and >1.5 cm; Strawberry tongue with cracked, red lips; Hands/feet, indurated, swollen, and red in the acute phase, with periungual desquamation in the convalescent phase (the second to third week) — the peeling occurs in convalescence, not the acute febrile phase, a direction the exam often flips. Treatment is IVIG plus high-dose aspirin, given within ten days of onset to lower the risk of coronary aneurysm. This is one of the rare pediatric situations calling for aspirin, and here it is for its antiplatelet and anti-inflammatory effects, not to reduce fever. An easy-point vaccine question: live attenuated vaccines (oral rotavirus, intradermal BCG, subcutaneous varicella, subcutaneous MMR) are contraindicated in pregnancy and severe immunodeficiency; the two injectable live vaccines, if not given on the same day, must be spaced at least four weeks apart, while inactivated vaccines carry no such restriction; live vaccines must be deferred for several months after recent IVIG or transfusion; and the two routes — intradermal BCG and oral rotavirus — are commonly mixed up. Mycoplasma pneumonia, lacking a cell wall, does not respond to β-lactams and requires a macrolide instead; diagnosis relies on cold agglutinins/IgM/PCR, since culture is extremely difficult.

To stress the ANA story once more: it is a screening tool for SLE, highly sensitive but not highly specific — low-titer positives are common in healthy people (about 20–30% at 1:40) — so specificity is not 90%, a common numeric trap. SLE is confirmed with dsDNA and Sm; anti-CCP is more specific for RA than RF and can predict it early; ANCA splits into c-ANCA (PR3, corresponding to GPA) and p-ANCA (MPO, corresponding to MPA and eosinophilic granulomatosis), both markers of vasculitis.

Graves' disease and Hashimoto's thyroiditis both belong to autoimmune thyroid disease, and the difference lies in whether the antibody "stimulates" or "destroys." Graves' key antibody is TRAb/TSI, which stimulates the TSH receptor → hyperthyroidism; that same TRAb also attacks the retro-orbital tissue, causing exophthalmos and glycosaminoglycan accumulation — so a single antibody explains two manifestations. Hashimoto's key antibodies are anti-TPO and anti-thyroglobulin, which slowly destroy the thyroid → mostly hypothyroidism, with pathology showing lymphocytic infiltration, germinal centers, and Hürthle cells. The trap lies here: a question describing "a thyroid biopsy showing dense lymphoplasmacytic infiltration, germinal centers, and Hürthle cells" is Hashimoto's pathology; Graves' pathology is "follicular epithelial hyperplasia with scalloped colloid vacuoles." What actually points a case toward Graves' is the combination of hyperthyroidism + exophthalmos + a positive TRAb — don't jump to Hashimoto's just because you see "lymphocytic infiltration."

Behçet's disease is a systemic vasculitis associated with HLA-B51, distributed along the old Silk Road. Its three cardinal features are recurrent oral ulcers (obligatory), genital ulcers, and ocular disease (uveitis). Keep "most common" and "most severe" straight for the uveitis: anterior uveitis is the most common (the traditional teaching, which can be accompanied by a hypopyon; large cohort studies actually find panuveitis most common, about 60%), whereas posterior uveitis, though more severe and capable of causing blindness, is not the most common. Another signature is a positive pathergy test — a pustule forming 24–48 hours after a needle prick.

Viral and bacterial questions look like two different mountains, but at their core they are the same range: map out structure and replication strategy first, and drug targets and resistance mechanisms will surface on their own. Where does a DNA virus replicate, is it double-stranded, does it use reverse transcription; can an RNA virus serve directly as mRNA, must it carry its own RdRp; is a bacterium's cell wall Gram-positive or Gram-negative, does it carry LPS, is it an intracellular organism. Set these pieces in place first, and every drug and resistance mechanism that follows is simply a different stop along this same route.

For nonenveloped DNA viruses, keep the mnemonic "bare PPPA": Parvo, Papilloma, Polyoma, Adeno — these four families lack an envelope; every other DNA virus has one. The core mechanism behind every oncogenic DNA virus is "inactivating tumor-suppressor genes": HPV's E6 degrades p53 and E7 inhibits pRb, and this pairing strips the cervical epithelium of its brakes; polyomavirus/SV40's large T antigen knocks out both p53 and pRb, while small t inhibits PP2A; adenovirus instead uses E1A binding pRb and E1B binding p53 — so the classic trap is pasting E6/E7 onto adenovirus, or swapping the targets of large T and small t; when you see the question, first sort out who is hitting what. EBV, HHV-8, and HBV each carry their own oncogenic role as well: EBV is linked to Burkitt lymphoma and nasopharyngeal carcinoma, HHV-8 is the culprit behind Kaposi sarcoma, and HBV's HBx protein interferes with p53 while the virus integrates into the host genome to cause hepatocellular carcinoma.

JC polyomavirus reactivates in the immunocompromised (AIDS, natalizumab use), primarily infecting oligodendrocytes and destroying myelin → progressive multifocal leukoencephalopathy (PML); it does not primarily infect neurons or microglia, and this directional point is frequently flipped in exam questions. One last iron rule: of the five hepatitis types, only B is a DNA virus, and it uses reverse transcription; A, C, D, and E are all RNA viruses.

Mapped onto representative families: positive-sense includes Picorna (enterovirus, HAV), Flavi (HCV, dengue, Japanese encephalitis), Corona, Toga (rubella), and Calici (HEV); negative-sense includes Orthomyxo (influenza), Paramyxo (measles, mumps, RSV), Rhabdo (rabies), Filo, Bunya, and Arena; dsRNA is Reovirus (rotavirus); reverse-transcribing is retrovirus (HIV) plus the special case of HBV.

The comparison across the five hepatitis types is a high-yield freebie. HAV is fecal-oral, non-chronic, and vaccine-preventable; HBV is a DNA virus, transmitted via blood and vertically from mother to child, and can become chronic (about 90% of infections acquired in infancy become chronic); HCV is positive-sense RNA, transmitted via blood, most prone to chronicity at roughly 80%, now curable with DAAs, and has no vaccine; HDV is a defective virus whose envelope borrows HBV's HBsAg, so it can only replicate inside HBV-positive hosts, either as coinfection or superinfection with HBV, and the HBV vaccine therefore also prevents HDV; HEV is fecal-oral and usually non-chronic (it can become chronic in immunocompromised hosts), but carries a high fatality rate in pregnant women. Quick HBV serology: HBsAg positive means current infection (chronic if it persists beyond six months), anti-HBs positive means recovery or vaccine immunity, anti-HBc IgM marks the acute phase or window period, and HBeAg positive indicates high infectivity and high viral replication.

The HIV life cycle maps one-to-one onto its drug targets. gp120 binds CD4 plus a coreceptor (CCR5/CXCR4), and gp41 mediates fusion into the cell — hence enfuvirtide, a 36-amino-acid peptide fusion inhibitor that binds gp41 HR1 and is given subcutaneously, while maraviroc blocks CCR5. Next comes reverse transcriptase (NRTIs such as zidovudine and lamivudine; NNRTIs) — lamivudine is also used for HBV; then integrase (the "-tegravir" drugs, such as dolutegravir); and finally protease (the "-navir" drugs, such as ritonavir). The trap is mistaking enfuvirtide for a nucleoside analog, or mistaking oseltamivir for an anti-HIV drug — oseltamivir is a neuraminidase inhibitor for influenza.

Influenza, rabies, and several other high-yield single points can likewise be deduced from structure. Influenza is negative-sense with eight gene segments — which is exactly why there is a distinction between "antigenic shift (gene reassortment causing pandemics)" and "antigenic drift (point mutations causing seasonal epidemics)"; HA handles attachment and entry, while NA cleaves sialic acid to release newly formed virions, so oseltamivir/zanamivir inhibit NA and block release and spread; the most common complications are secondary pneumonia from S. pneumoniae and S. aureus. LAIV (the live attenuated intranasal vaccine) is approved only for healthy people aged 2 to 49; those over 50, pregnant women, the immunocompromised, and children under 2 are all ineligible and should instead receive the inactivated vaccine. Rabies belongs to Rhabdoviridae, a bullet-shaped negative-sense RNA virus, and Negri bodies (eosinophilic inclusions) are seen in the neuronal cytoplasm; the leading cause of death in EV71 is brainstem encephalitis, producing neurogenic pulmonary edema and cardiopulmonary failure — not intestinal electrolyte loss and not latent infection — and ADE (antibody-dependent enhancement) belongs to dengue, not EV71. Rubella is diagnosed by serum IgM or RT-PCR, not by routine throat-swab culture.

The three pillars of Clostridium must be kept straight. C. perfringens' α-toxin is a lecithinase that dissolves cell membranes, causing muscle necrosis with gas production (crepitus) — this is gas gangrene, not necrotizing fasciitis; necrotizing fasciitis (the "flesh-eating" infection) is Streptococcus pyogenes tracking along the fascial planes, and the two are the pair most often confused. C. difficile's toxins A and B cause pseudomembranous colitis, triggered by the "3 C's" of prolonged broad-spectrum antibiotics — clindamycin, cephalosporins, fluoroquinolones — which disrupt the normal flora; clindamycin and ampicillin are triggers, not treatments; treatment means stopping the inciting antibiotic and switching to oral vancomycin or fidaxomicin (the 2021 IDSA guidelines have already demoted metronidazole to a second-line option). C. tetani's tetanospasmin blocks the release of glycine/GABA from inhibitory spinal interneurons, causing disinhibition and spastic paralysis; C. botulinum blocks ACh release at peripheral motor nerve terminals, causing flaccid paralysis — same genus, opposite mechanisms; do not mix up the direction.

To read the logic of tuberculosis diagnostic tools, first ask whether the question is asking "has this person ever been infected (immune memory)" or "are there live organisms right now." TST (the skin test) reads a delayed-type hypersensitivity reaction and is confounded by BCG and NTM, producing false positives; IGRA uses ESAT-6 and CFP-10, two antigens encoded by the RD1 gene region, to stimulate T cells and measure IFN-γ, and because BCG has had RD1 deleted during its preparation, a person who received BCG can still test IGRA-negative, making it more specific than TST; but neither test can distinguish latent from active tuberculosis. AFB smear is fast and cheap but has low sensitivity, requiring roughly 10⁴ organisms/mL, and cannot distinguish TB from NTM. NAAT/PCR can detect tuberculosis DNA within hours and simultaneously test for rifampin resistance (rpoB), but cannot fully characterize the complete resistance profile — the claim that "DNA-based technology can fully characterize drug resistance" is false. Culture (MGIT), though slow, remains the gold standard and the basis for a complete susceptibility panel.

Immunity to tuberculosis is Th1, not Th2: macrophages present antigen → Th1 secretes IFN-γ and TNF-α → macrophages are activated and form granulomas (epithelioid cells plus Langhans giant cells). Th2 handles parasites and allergy (IL-4, IL-5, IgE), so forcing tuberculosis onto the Th2 pathway is simply wrong. Tick-borne intracellular organisms share one treatment iron rule: Ehrlichia, Anaplasma, and Rickettsia are all first treated with doxycycline. Ehrlichia chaffeensis invades monocytes (morulae can be seen), while Anaplasma phagocytophilum invades neutrophils — swapping the two is a common trap. Francisella tularensis is the exception, where severe cases may switch to streptomycin/gentamicin.

β-lactams strike the PBPs and the D-Ala-D-Ala terminus; vancomycin and dalbavancin bind D-Ala-D-Ala. MRSA's resistance comes from PBP target alteration (switching to PBP2a), so not even carbapenems can bind, and adding a β-lactamase inhibitor is useless — because the drug is not being cleaved by an enzyme, the target itself has changed. VRE's resistance comes from changing the cell-wall terminus from D-Ala-D-Ala to D-Ala-D-Lac, dropping the binding affinity of vancomycin and dalbavancin roughly a thousandfold, so both become ineffective. Fluoroquinolone's targets are GyrA and ParC, and mutations at these two sites confer FQ resistance. Sulfonamides strike folate-synthesizing enzymes, entirely unrelated to PBPs, so the claim that "PBP mutation affects sulfonamides" is wrong; Gram-negative organisms also do not resist via endospore formation — endospores belong only to the Gram-positive Bacillus and Clostridium. Linezolid blocks the 50S initiation complex, not the cell wall.

Within the Ambler classification of β-lactamases, class B is a metalloenzyme (MBL) that depends on Zn²⁺ and can hydrolyze carbapenems — exactly what makes the global spread of NDM, IMP, and VIM so alarming, since they are unaffected by clavulanate and require specialized inhibitors. The trap examiners bury most often is the statement "class B cannot break down carbapenems" — the exact opposite is true. Class A includes ESBL/KPC (of which KPC can break down carbapenems), class C is AmpC, and class D is OXA. β-lactamase inhibitors (clavulanate, tazobactam; sulbactam is the exception — it binds PBPs and is active on its own against Acinetobacter) have almost no bactericidal activity of their own; they simply bind irreversibly to and inhibit β-lactamase to protect the co-administered β-lactam — so they are entirely useless against MRSA, because MRSA resistance is target alteration, not enzymatic cleavage.

The first-line tuberculosis regimen HRZE uses four drugs for intensive treatment during the first two months, then H plus R for continuation over the following four months. Isoniazid inhibits mycolic acid synthesis, with toxicities of peripheral neuropathy (supplement B6) and hepatotoxicity; rifampin inhibits RNA polymerase, turning body fluids orange-red, causing hepatotoxicity, and inducing CYP enzymes; pyrazinamide acts in an acidic environment, causing hyperuricemia and hepatotoxicity; ethambutol inhibits arabinosyl transferase, and the toxicity most worth remembering is optic neuritis with red-green color blindness — do not confuse this with INH's peripheral neuropathy.

The three segments of LPS, from outside to inside, are the O antigen (the outermost polysaccharide chain, used for serotyping, as in O157) → the core polysaccharide → lipid A (the active core of endotoxin: fever, hypotension, DIC, and complement activation with macrophage release of TNF and IL-1). The O antigen comes from the LPS outer polysaccharide chain, the H antigen from flagellar protein (flagellin), and the K antigen from the capsular polysaccharide — do not confuse the source of these three. Porins sit in the outer membrane (not the inner membrane), and loss or alteration of porins is one mechanism of resistance (the drug can no longer get in).

Quick recall of representative organisms likewise grows out of mechanism. Salmonella Typhi is fecal-oral and spreads person-to-person (not directly from poultry or livestock), with human carriers residing in the gallbladder; non-typhoidal Salmonella comes from eggs, meat, and salads, producing delayed-onset diarrhea without blood. A positive Kanagawa test for Vibrio parahaemolyticus indicates production of TDH (thermostable direct hemolysin), the basis of its virulence; Vibrio cholerae relies on cholera toxin activating Gs to produce massive rice-water diarrhea. Legionella is a facultative intracellular organism cleared by cell-mediated immunity (macrophages), not by anticapsular antibody; it grows on BCYE medium. The distinctive feature of Yersinia pestis plague is that the flea's proventriculus becomes blocked with organisms → biting triggers regurgitation that injects the organism; a question that says "vomiting" is wrong. Shigella is fecal-oral and infectious at an extremely low inoculum, with Shiga toxin producing bloody stools and dysentery; EHEC (O157:H7/STEC) comes from undercooked beef, causing hemorrhagic colitis and HUS in children.

How do bacteria pass resistance genes to one another? Three routes: transformation, taking up free DNA (requires natural competence), transduction, a bacteriophage carrying bacterial DNA, and conjugation, a sex pilus transferring a plasmid directly — the last of these is the main route by which resistance plasmids spread. A plasmid is extrachromosomal circular dsDNA that replicates independently and occurs in both Gram-positive and Gram-negative organisms, often carrying resistance genes such as β-lactamase. Mutation types are classified as silent (base changes, amino acid unchanged), missense (swapped for a different amino acid), nonsense (changed to a stop codon, prematurely truncating the protein), and frameshift (the entire reading frame is scrambled); do not confuse silent with missense. The lac operon has dual control: glucose holds the master switch (catabolite repression, cAMP/CAP), while lactose releases the brake (the repressor is displaced by allolactose); quorum sensing regulates genes according to cell density and has nothing to do with the lac operon.

In A-B exotoxins, the B subunit handles binding and delivery while the A subunit carries the enzymatic activity: diphtheria toxin's A subunit ADP-ribosylates and inactivates EF-2, blocking protein synthesis; Pseudomonas exotoxin A likewise inactivates EF-2; cholera toxin activates Gs, raising cAMP to produce massive watery diarrhea; Shiga toxin cleaves 28S rRNA, inhibiting protein synthesis. The trap is writing diphtheria toxin's target as a ribosomal subunit or as "EF-3" (EF-3 does not exist in human cells). Mycobacterium tuberculosis survives intracellularly by blocking phagosome-lysosome fusion; its cell wall is rich in mycolic acid, which accounts for its acid-fast positivity, its GC content of roughly 65%, on the high side, and its cell membrane, which contains no sterols. A prion is fundamentally a misfolded protein, PrPSc, that contains no nucleic acid, so UV and ionizing radiation are nearly useless against it; inactivation requires prolonged high-temperature, high-pressure treatment plus NaOH or sodium hypochlorite.

Gram staining earns its title as "the first cut of triage" because it splits the bacterial world cleanly in two: organisms with a thick cell wall, multilayered peptidoglycan, and teichoic acid take up the purple stain and are Gram-positive; those with a thin wall and an outer membrane cannot hold the stain and are counterstained red with safranin. This section first clears out the Gram-positive family tree, then walks into the fungi, and finally follows the CD4 count downstairs — and you will find that all three seemingly different topics in this chapter are really asking the same thing: once immunity collapses, who comes running out?

Subgrouping the β-hemolytic streptococci is another freebie, as long as you remember the causal chain: group A S. pyogenes is bacitracin-sensitive and PYR-positive; group B S. agalactiae is bacitracin-resistant, CAMP-test positive, and hydrolyzes hippurate, and it is the culprit behind neonatal sepsis and meningitis. On the CoNS side: think of S. saprophyticus, resistant to novobiocin, for urinary tract infection in a young woman, while S. epidermidis is, conversely, sensitive.

S. aureus is so formidable because it carries a complete toolkit for "sticking fast, breaking free, and slipping past" the immune system. Protein A binds the Fc portion of IgG backward, effectively wearing the antibody as a cape turned inside out, so macrophage Fc receptors cannot find their target and opsonophagocytosis is blocked. Coagulase cleaves fibrinogen into fibrin, wrapping the bacterium inside a clot to hide from immunity; but to spread, it must reverse the operation, so staphylokinase emerges to dissolve that fibrin — the same organism forms a clot with one hand to protect itself and dissolves it with the other to clear a path. The exam loves to reverse these two directions: writing that staphylokinase "forms" fibrin is wrong. Enterotoxin is heat-stable, and ingesting the preformed toxin produces vomiting after an extremely short incubation period; TSST-1 is a superantigen that bypasses antigen specificity entirely to drag out massive numbers of T cells and detonate a cytokine storm — this is toxic shock syndrome.

The test points on Gram-positive bacilli concentrate on three organisms, and each one's signature trait traces back to a clear causal chain. Corynebacterium diphtheriae infects only humans, with no animal reservoir, and the gene for the toxin that actually kills is carried on a bacteriophage β — not a plasmid; getting this location backward is simply wrong. The toxin itself works by ADP-ribosylating EF-2, locking up the elongation factor for protein synthesis, so mucosal cells die and pile up into that grayish-white pseudomembrane that cannot be peeled away; Pseudomonas exotoxin A shares the same mechanism. Bacillus anthracis has a capsule made of D-glutamate and forms spores; its three-component toxin is PA + EF + LF: PA is the key that lets it into the cell, EF is a calmodulin-dependent adenylate cyclase that raises cAMP and causes edema, and LF is the lethal factor. Remember that EF belongs to anthrax, not to S. aureus; anthrax classically presents in cutaneous, inhalational, and gastrointestinal forms (CDC now also lists injection anthrax, seen in people who inject drugs), and there is no "muscle-necrosis form" — cross out any option describing muscle necrosis on sight.

Listeria monocytogenes is a Gram-positive facultative intracellular organism — note that it is "facultative," not obligate, a word the exam often swaps out on you. It has two signature traits: first, it still grows under refrigeration at 4°C, which is why chilled soft cheeses and deli meats serve as sources of infection; second, it grows flagella and performs "tumbling motility" at 25°C, yet loses motility at 37°C. Once inside a cell, it propels itself by polymerizing actin via ActA, rocketing between cells to evade antibody attack — which is also why pregnant women, neonates, the elderly, and the immunocompromised progress to meningitis and sepsis once infected.

One last point to close with — small, but frequently flipped on the exam: the pneumococcal vaccine's component is capsular polysaccharide, not a surface protein. PPSV23 is pure polysaccharide, while PCV13/15/20 conjugate the polysaccharide to a protein, recruiting T cells and inducing a memory response, which is why it suits infants whose immune systems are not yet mature. While we are at it, let's clear up two toxins often misassigned: the heat-stable enterotoxins STa/STb belong to enterotoxigenic E. coli (ETEC), not to S. pyogenes; and edema factor belongs to anthrax, not to S. aureus.

Nearly every test point in the world of fungi revolves around "what shape does it take." First, remember one thing: the fungal cell membrane replaces cholesterol with ergosterol, and this is the principal target of antifungal drugs; the cell wall, meanwhile, is chitin plus β-glucan. Morphology falls into four broad categories: unicellular budding yeasts; hyphae with cross-walls, called septate hyphae; hyphae without cross-walls and with broad, wide lumens, called coenocytic hyphae; and dimorphic fungi, which are molds at 25°C and switch to yeast at 37°C.

Among the dimorphic fungi there is one favorite exception: at 37°C in tissue, Coccidioides immitis does not turn into a yeast but into a spherule, packed with large numbers of endospores that spray outward and cause dissemination when it ruptures; the other dimorphic fungi (Histoplasma, Blastomyces) are the ones that actually turn into yeast at 37°C. Histoplasma capsulatum's signature is small yeast forms inside macrophages, sourced from bird droppings, bat guano, and the Mississippi River Valley; Blastomyces is a large yeast with broad-based budding.

A handful of unusual fungi also come up often. Malassezia furfur is lipid-dependent, so the culture medium must be supplemented with olive oil for it to grow; the "spaghetti and meatballs" pattern of short hyphae plus round spores seen on skin scrapings is this organism, and it causes tinea versicolor. Candida albicans sprouts a germ tube at 37°C or in serum, which is the basis of a rapid identification test; Candida glabrata is the exception — it exists only as a unicellular yeast and forms no pseudohyphae or hyphae — and it is frequently resistant to azoles. Cryptococcus neoformans has a thick capsule, and on Niger seed (birdseed) agar, phenol oxidase oxidizes the substrate into melanin, giving the colonies a coffee-brown color; diagnosis relies on India ink staining or CrAg. C. gattii is its twin, differing in that it infects immunocompetent hosts yet produces more severe neurological disease — do not be fooled by "immunocompetent" into thinking it is milder.

The spore distribution across the three dermatophyte genera is another point that is wrong the moment you flip it: Epidermophyton produces no microconidia, only club-shaped macroconidia; Microsporum is rich in macroconidia (spindle-shaped, thick-walled); Trichophyton is rich in microconidia.

The test point for antifungal drugs is always "direction." Polyenes (amphotericin B, nystatin) bind ergosterol directly, punching holes in the membrane that let K⁺ leak out — this is direct binding, not inhibition of synthesis, and this direction is the licensing exam's favorite point to flip. Azoles (fluconazole, etc.) inhibit 14-α-demethylase, inhibiting ergosterol synthesis. Terbinafine inhibits squalene epoxidase, also upstream in the synthetic pathway. Echinocandins (caspofungin) strike not the membrane but the β-glucan synthase of the cell wall. 5-Flucytosine inhibits nucleic acid synthesis and is combined with amphotericin B to treat cryptococcal meningitis.

HIV can drive a person toward death because it turns CD4+ T cells, the very hub of cell-mediated immunity, into its own replication factory.

The CD4 count is itself a timer: as the tide recedes, the reefs lying in wait beneath the surface surface one by one.

The core of any AIDS-defining-illness question is simply "is it on the list." Kaposi sarcoma, non-Hodgkin lymphoma (including Burkitt, primary CNS lymphoma, and immunoblastic lymphoma), and invasive cervical cancer are all on it; so are PJP, esophageal candidiasis, CMV retinitis, Toxo encephalitis, Cryptococcus meningitis, disseminated MAC, and extrapulmonary or disseminated TB. Note: although the incidence of Hodgkin lymphoma rises in HIV patients, it is not AIDS-defining — this is a high-frequency trap. A CD4 count below 200 alone already meets the criterion for AIDS.

Do not mix up the numbers for mother-to-child transmission either: roughly 25–30% without treatment, falling to under 1% when complete antiretroviral therapy during pregnancy suppresses the viral load to undetectable — this is the clinical basis of "U=U" (undetectable = untransmittable). The TB/HIV coinfection question runs finer: treat TB first; do not start ART immediately alongside it, to avoid immune reconstitution inflammatory syndrome (IRIS). The specific timing splits as follows: when CD4 is below 50, start ART within 2 weeks of beginning anti-tuberculosis therapy, because the mortality risk outweighs the risk of IRIS; when CD4 is 50 or above (say, 70), delay to 2 to 8 weeks; for tuberculous meningitis, early ART is inadvisable regardless of CD4 count and should be deferred to 4 to 8 weeks after starting anti-tuberculosis therapy, because starting early actually raises mortality. Another drug-interaction trap: rifampin is a potent CYP3A4 inducer that substantially lowers blood levels of protease inhibitors (such as lopinavir/ritonavir), so patients on a PI who need treatment for latent TB must not use rifampin — switch instead to rifabutin (dose-adjusted) or nine months of isoniazid alone.

Opportunistic infections in transplant recipients or others with cell-mediated immune deficiency also follow a timeline: within 1 month post-transplant, most are surgery-related or represent reactivation of latent donor or recipient infection (early HSV, CMV); from 1 to 6 months, immunosuppression is at its deepest, and CMV, PJP, and BK virus all arrive together; after 6 months, the picture largely returns to community-acquired infections. On seeing a combination like "recurrent herpes zoster plus Listeria meningitis," immediately localize it to a cell-mediated (T-cell) immune defect, and the next step is a blood count plus lymphocyte phenotyping for CD4/CD8.

The Surviving Sepsis Campaign packages what must be done in the first hour into a bundle, and every step has its own logic. Step one is measuring lactate, which reflects perfusion; step two is drawing blood cultures before antibiotics, since giving the drug first would compromise the culture; step three is giving broad-spectrum empiric antibiotics within one hour, as mortality rises with every hour of delay; when there is hypotension or a lactate of 4 or higher, step four is a rapid infusion of 30 mL/kg of crystalloid to restore preload; if blood pressure remains low after fluids, step five is starting a vasopressor, targeting a MAP of at least 65. The first-line vasopressor is norepinephrine, with vasopressin added if insufficient, and hydrocortisone considered for refractory cases. Sodium bicarbonate is not a priority — the real fix lies in fluids, antibiotics, and vasopressors; bicarbonate is generally reserved for specific situations such as a pH below 7.2.

Matching the infection source to the likely pathogen is also frequently tested.

Two points are frequently reversed on the exam: the main cause of SBP is Gram-negative enteric organisms, not Gram-positive ones — in a cirrhotic patient the intestinal mucosal barrier collapses and gut bacteria translocate into the ascitic fluid, so enteric organisms naturally dominate; and tissue-culture positivity in cellulitis runs only 20–30%, so diagnosis is almost always clinical — inflating that figure to 70% is wrong.

The pairing of organism to clinical setting all makes causal sense: subacute disease after a dental procedure — the culprit is viridans streptococci, oral residents with strong adhesion; tricuspid-valve disease in IVDU — S. aureus is the fiercest and most common; colorectal cancer or a colonic lesion — S. gallolyticus (formerly bovis), and finding it means a colonoscopy to look for cancer; early prosthetic-valve infection — S. epidermidis settles in via biofilm; culture-negative disease — think of slow-growing or intracellular organisms such as HACEK, Coxiella (Q fever), and Bartonella, which require special culture techniques or serology. Enterococcus favors elderly patients after urologic or gastrointestinal procedures, and treatment requires a cell-wall-active agent combined synergistically with an aminoglycoside.

The modified Duke criteria require 2 major criteria, or 1 major plus 3 minor, or 5 minor, to confirm IE. There are only two major criteria: positive blood cultures with a typical organism (viridans strep, S. gallolyticus, HACEK, community-acquired S. aureus, or Enterococcus, positive on two separately drawn sets, or persistent bacteremia); and evidence of endocardial involvement (echocardiography showing an oscillating vegetation, an abscess, prosthetic-valve dehiscence, or new valvular regurgitation). Fever, Janeway lesions, and Osler nodes are all minor criteria, not major — this is the most frequently tested trap.

The first-line imaging study is transthoracic echocardiography (TTE), but transesophageal echocardiography (TEE) is required when the valve is prosthetic or when TTE is negative yet suspicion remains high, since it offers higher sensitivity. The 2023 Duke-ISCVID update adds direct visualization at surgery as a new major criterion, adds PET-CT (¹⁸F-FDG) and cardiac CT to the imaging major criterion (particularly useful for prosthetic-valve and device infection), adds PCR/pathogen gene sequencing and the Bartonella enzyme immunoassay to the microbiologic major criterion, drops the rigid requirement for "two separately drawn sets with a time interval," and adds TAVI and cardiac implantable electronic devices (CIEDs) as new risk factors. The core concept is unchanged: the major criteria are typical-organism bacteremia plus evidence of endocardial involvement, while the peripheral stigmata remain minor.

For treatment, empiric drug choice branches by setting: acute native-valve disease must cover S. aureus (including MRSA → vancomycin); a prosthetic valve is treated with vancomycin + gentamicin + rifampin, then de-escalated once the organism is identified. The course is typically 4 to 6 weeks of intravenous therapy, because the vegetation has no blood supply and drug penetration is poor. The three major surgical indications also each have a traceable cause: heart failure from valve destruction or a ruptured chordae causing acute severe regurgitation — this is the most common and most important surgical indication; uncontrolled infection, meaning persistent bacteremia beyond 7 to 10 days despite appropriate antibiotics, a perivalvular abscess, conduction block, or a resistant organism or fungus; and prevention of embolism, meaning a large vegetation (>10 mm) with an embolic event, or recurrent IE on a prosthetic valve. Still having fever or a positive culture in the first 3 days of effective antibiotic therapy is a normal response, not a surgical emergency — the true threshold of persistent bacteremia that calls for surgery is 7 to 10 days.

The scope of antibiotic prophylaxis has also been sharply narrowed: a single preoperative dose of amoxicillin (clindamycin/azithromycin for the penicillin-allergic) is needed only for the highest-risk cardiac conditions (a prosthetic heart valve, a prior episode of IE, unrepaired or incompletely repaired cyanotic congenital heart disease, or valvulopathy after heart transplantation) combined with a dental procedure that breaches the gingiva, the periapical region, or the oral mucosa. Six months after cyanotic congenital heart disease is completely repaired with prosthetic material, endothelialization is complete, the risk returns to that of the general population, and prophylaxis is no longer needed. Simple MVP, rheumatic heart disease, and ASD do not require dental prophylaxis; gastrointestinal or genitourinary procedures are no longer routinely recommended for prophylaxis.

Travel-related infection questions are almost all reflex questions of "exposure → pathogen → treatment" — memorize four main threads and you win half the battle. Southeast Asia or Thailand + diabetes + pulmonary infiltrates + a splenic abscess + a Gram-negative bacillus with a "safety-pin" appearance (staining darker at both ends, paler in the middle) — this is Burkholderia pseudomallei (melioidosis), and diabetes is its single most important risk factor; treatment is ceftazidime during the intensive phase, or meropenem for severe disease, for at least 10 to 14 days, followed by an eradication phase of oral TMP-SMX for 3 to 6 months to prevent relapse. Pneumonia after a cruise or hotel stay + diarrhea + altered mentation + hyponatremia — Legionella pneumophila (Legionnaires' disease), sourced from cooling towers or water systems, diagnosed by urinary antigen. Wading through water or flooding, rat-urine contamination + fever and myalgia + conjunctival injection ± jaundice and renal failure (Weil's disease) — Leptospira, and note that this is transmitted by contact, not by mosquitoes, entering through skin or mucosa. A generalized rash involving the palms and soles after sexual contact + condyloma latum — secondary syphilis.

The test point for traveler's diarrhea is geography. It is mostly ETEC, but Campylobacter in Southeast Asia carries a high rate of fluoroquinolone resistance, so the first-line choice switches to azithromycin for Southeast Asia, pregnant women, and children. Bismuth subsalicylate can be used for prevention (blocking 50–60%), but it contains salicylate → contraindicated in children (Reye syndrome risk). A pregnant woman with suspected influenza should be given oseltamivir and must not have it delayed because of pregnancy — "antivirals are not recommended in pregnancy" is false; antiviral therapy has the greatest benefit within 48 hours of symptom onset, and is still recommended for high-risk groups even beyond that window. The most common cause of the common cold is rhinovirus, with over a hundred serotypes and no effective vaccine.

Syphilis treatment carries one more classic trap: the Jarisch-Herxheimer reaction — once the spirochetes are killed, a burst of released antigen triggers acute inflammation (fever, chills, headache, myalgia). It occurs most often after treating secondary syphilis (roughly 50–90%), never under 15%; it usually resolves on its own within 12 to 24 hours, treatment does not need to be stopped, and the drug remains benzathine penicillin G.

Classic FUO (the Petersdorf definition): a temperature repeatedly above 38.3°C, a course of 3 weeks or more, with no cause found after 1 week of inpatient workup or three outpatient visits. Summed up in one line: prolonged, high fever with no answer. The three major causal categories, ranked by weight, are: infection as the largest category (tuberculosis, abscess, IE, EBV/CMV); tumor next (lymphoma, leukemia, renal cell carcinoma); and connective-tissue/autoimmune disease (adult-onset Still disease, giant cell arteritis, SLE). The order of reasoning is history-taking for travel and exposures → physical exam (lymph nodes, cardiac murmurs, liver and spleen) → basic bloodwork plus cultures plus imaging → biopsy or PET if needed — the cardinal sin is firing antibiotics blindly from the start and masking the cultures.

The classic triad of EBV infectious mononucleosis is fever + exudative pharyngitis + posterior cervical lymphadenopathy, often with splenomegaly; diagnosis relies on atypical lymphocytes plus a Monospot (heterophile antibody) test, with EBV-VCA IgM for confirmation. Two traps: mistakenly giving amoxicillin/ampicillin triggers a diffuse rash (not a true allergy); and splenomegaly means avoiding contact sports (risk of splenic rupture). The incubation period is 30 to 50 days, not 5 to 10; an unvaccinated person exposed to varicella has an infection rate above 90%, reflecting its extreme contagiousness. CMV can also cause a similar mononucleosis-like illness with atypical lymphocytes, but with a negative heterophile antibody, which distinguishes it from EBV.

Ludwig's angina is a rapidly progressive cellulitis of the bilateral submandibular and submental spaces, most often arising from a lower-molar dental infection. The lethal mechanism is swelling of the floor of the mouth that pushes the tongue base up and back, obstructing the airway; management is securing the airway (awake intubation or tracheostomy if needed) + intravenous broad-spectrum antibiotics + surgical drainage — observation alone is not acceptable. Rhino-cerebral fungal infection favors patients with DKA or immunosuppression, presenting with a black necrotic eschar of the nasal palate plus orbital cellulitis or cavernous sinus invasion; histologic sorting: Mucor is non-septate, broad, with 90° right-angle branching; Aspergillus is septate, with 45° acute-angle branching — treatment for both is surgical debridement plus amphotericin B along with correcting the underlying predisposing factor. Osteomyelitis in adults is most commonly caused by S. aureus, while patients with sickle cell disease are predisposed to Salmonella; the diagnostic sequence is plain X-ray first (though often still normal within 2 weeks) → MRI, the most sensitive for marrow edema and soft-tissue involvement; bone biopsy culture is the gold standard for identifying the pathogen.

The key variable in empiric antibiotics is Listeria. The standard coverage for a healthy adult with suspected bacterial meningitis is vancomycin + a third-generation cephalosporin (ceftriaxone or cefotaxime), covering S. pneumoniae and N. meningitidis. But in a neonate, someone over 50, a pregnant woman, or an immunocompromised patient, you must add ampicillin to cover Listeria — because cephalosporins do not cover Listeria. Listeria is a Gram-positive bacillus (not a coccus, not a spore-former), and ampicillin is first-line, with an aminoglycoside added in the immunocompromised. Vancomycin and ciprofloxacin both perform poorly against Listeria and are not first-line. At the same time, do not forget to give dexamethasone before or together with the first dose of antibiotics, which reduces hearing loss and neurologic sequelae in pneumococcal meningitis.

Cryptococcal meningitis favors patients with HIV/AIDS (CD4 < 100); CSF India ink staining shows the capsule, and CrAg is positive. The course runs in three phases, not four weeks: an induction phase of amphotericin B + flucytosine for at least 2 weeks; a consolidation phase of fluconazole for 8 weeks; and a maintenance phase of low-dose fluconazole for at least 1 year or until CD4 recovers. The total course exceeds 12 weeks, and elevated intracranial pressure must also be managed (serial lumbar drainage).

Tuberculous meningitis: CSF shows a lymphocyte predominance with low glucose and high protein, classic basal meningeal enhancement, and it can be complicated by hydrocephalus or cranial nerve palsies. Culture remains the gold standard (though it takes 4 to 6 weeks); PCR/Xpert improves early detection but has not replaced culture — the claim that "PCR has replaced culture" is false.

The single most important thing to remember about HSV encephalitis is that "treatment cannot wait." HSV-1 invades the temporal lobe, producing fever, personality and behavioral change, aphasia, and focal seizures, with MRI/EEG showing temporal-lobe lesions. CSF HSV PCR has a sensitivity and specificity above 95%, but it can be falsely negative within the first 72 hours of illness, falling only as the disease progresses further. When it is suspected, give IV acyclovir first and do not wait for test results — delay carries a high mortality. Another trap: PCR often remains positive for several days of treatment — the claim that "PCR positivity drops below 10% within 3 days" is false; PCR can still be positive 7 to 14 days into treatment.

The signs of meningeal irritation must also be kept straight: Brudzinski's sign is involuntary flexion of the hips and knees triggered by passive neck flexion; Kernig's sign is the inability to extend the knee, because of pain or resistance, after flexing the hip to 90 degrees; and nuchal rigidity is increased resistance to passive forward flexion of the neck.

Hand hygiene has two situations in which alcohol-based hand rub fails: visibly soiled hands (organic matter blocks the alcohol) and after caring for a patient with C. difficile (C. diff spores resist alcohol) — in both situations, soap and running water must be used for physical scrubbing.

Febrile neutropenia is defined as an ANC < 500 (or an ANC expected to fall below 500 within 48 hours) plus a single temperature ≥38.3°C, or ≥38.0°C sustained for 1 hour. Coverage must include Pseudomonas aeruginosa, because Gram-negative bacteremia is lethal; first-line therapy is a single antipseudomonal β-lactam: cefepime, piperacillin-tazobactam, or a carbapenem (imipenem/meropenem). Vancomycin is not first-line — it is added only for suspected catheter infection, skin/soft-tissue infection, hemodynamic instability, or known MRSA colonization.

MRSA, because mecA encodes PBP2a, is resistant to all β-lactams (including cephalosporins, oxacillin, and nafcillin) (exception: the newer-generation ceftaroline). Effective agents include vancomycin (first-line), teicoplanin, linezolid, daptomycin (never for pneumonia, since it is inactivated by pulmonary surfactant), and tigecycline. Institutional control measures: contact precautions plus active screening plus hand hygiene.

Get the memory right for the question on renal dose adjustment of antibiotics: metronidazole, azithromycin, ceftriaxone (dual elimination), moxifloxacin, linezolid, clindamycin, and doxycycline do not need renal adjustment; vancomycin, aminoglycosides, levofloxacin, cefazolin and most other β-lactams, and acyclovir do need adjustment. Metronidazole is mainly metabolized by the liver, so it is not reduced in renal failure (though it should be reduced in severe liver disease).

Metal chelation: tetracycline/doxycycline and fluoroquinolones (such as ciprofloxacin) chelate with divalent and trivalent metal ions (Ca²⁺, Mg²⁺, Al³⁺, Fe²⁺) to form non-absorbable complexes, so they should be avoided with milk, Mg/Al-containing antacids, iron supplements, and calcium tablets, and dosing should be separated by 2 hours.

When antibiotics should not be prescribed: purulent nasal discharge in a viral URI is not a reliable indicator of bacterial infection (viruses can produce it too), and antibiotics should not be given immediately in the absence of fever, sinus tenderness, or other local signs of bacterial infection; asymptomatic bacteriuria (except in pregnancy or before a urologic procedure) is not treated.

Needlestick exposure: irrigate immediately, assess the source, and start HIV PEP as soon as possible (ideally within 2 hours, at the latest within 72 hours), for a 4-week course. A needlestick is an emergency, and even on New Year's Eve the patient should go to the emergency department that same night — "delaying it to an outpatient visit after the holiday" is the classic correct answer for "the most inappropriate management."

Whether upper or lower, the most common causative organism in both is E. coli (75 to 85% of community-acquired cases), followed by Klebsiella, Proteus, and, in young women, S. saprophyticus. Empiric first-line therapy: nitrofurantoin/TMP-SMX for simple cystitis; a fluoroquinolone or a third-generation cephalosporin for pyelonephritis.

Sorting neuropathic bladder is also frequently reversed on the exam: cervical or upper spinal cord injury (with S2-S4 intact) → a spastic (UMN) bladder, with detrusor hyperreflexia, frequent involuntary contractions, and small capacity; injury at or below the S2-S4 micturition center (poliomyelitis, myelodysplasia, cauda equina, or sacral cord lesions) → a flaccid (LMN) bladder, with an areflexic detrusor, an overdistended bladder, and overflow incontinence. Cervical spinal injury produces a spastic UMN bladder, not a flaccid one — this is a high-frequency point that gets reversed.

The imaging pathway for recurrent UTI in children: renal ultrasound comes first (looking for hydronephrosis and anatomic abnormality, with no radiation); followed by a VCUG to rule out vesicoureteral reflux (VUR) (the most common anatomic abnormality behind recurrent infant UTI, graded I through V); a DMSA scan assesses renal scarring or acute pyelonephritis; and a diuretic renogram (DTPA/MAG3) assesses obstruction. For an infant with recurrent UTI plus mild hydronephrosis, start with ultrasound, then prioritize a VCUG — CT, IVP, and MRI are all not first-line.

UPJ obstruction: fever plus CVA tenderness plus severe hydronephrosis without ureteral dilation, and no VUR on VCUG; diagnosis relies on DTPA/MAG3 showing delayed drainage; the operation is pyeloplasty, not ureteral reimplantation (reimplantation is for VUR).

Interstitial cystitis (IC) is a non-infectious, non-bacterial chronic bladder pain syndrome — the urinalysis is usually normal or only mildly abnormal, with no pyuria and no bacteria; the complaint is bladder pain and pressure, frequency, and nocturia, worsening as the bladder fills and easing after voiding — do not mistake it for infection and prescribe antibiotics. The most common cause of unresolved bacteriuria (bacteriuria persisting after completing the first course of antibiotics) is bacterial resistance, with inadequate dosing, mixed infection, and poor adherence as other causes. The most common cause of hematuria in children (excluding neonates) is glomerular (although UTI is the most common cause of gross hematuria overall) — APSGN or IgA nephropathy, with tea- or cola-colored urine plus proteinuria plus RBC casts; painless gross hematuria in an adult should raise concern for a tumor; and BPH can also cause hematuria through mucosal vascular engorgement.

The terminal complement components C5 through C9 form the membrane attack complex (MAC), which punches holes in the bacterial membrane to lyse it. Neisseria species have a thin cell wall and are especially dependent on MAC for clearance — so a deficiency of terminal complement (C5-C9) → recurrent N. meningitidis or N. gonorrhoeae infection. The clue is recurrent meningococcemia or disseminated gonococcal infection in a young person, prompting a workup with CH50, which will be abnormally low, and an abnormal AH50; deficiency of the early complement components C1 through C4 leads toward an SLE-like picture or encapsulated-organism infection, while C3 deficiency leads toward severe, recurrent pyogenic infections. Vaccination against Neisseria meningitidis is an important preventive measure.

Organophosphate poisoning switches the parasympathetic system "fully on" — organophosphates irreversibly inhibit acetylcholinesterase, ACh accumulates massively, and both muscarinic and nicotinic receptors are persistently overactivated. The muscarinic side is captured by DUMBELS: Diarrhea, Urination, Miosis, Bradycardia, Emesis, Lacrimation, Salivation — increased secretions and pinpoint pupils; the nicotinic side is fasciculations, muscle weakness, and tachycardia; and the CNS side is altered consciousness, seizures, and respiratory depression. "Decreased salivation and tearing, with dilated pupils" points in the anticholinergic direction — the exact opposite. Treatment has three pillars: atropine to block muscarinic receptors, pralidoxime (2-PAM) to reactivate cholinesterase (which must happen before aging occurs), and activated charcoal for decontamination; physostigmine is an absolute contraindication — it is itself a cholinesterase inhibitor and would worsen the crisis.

Thunderclap headache — a headache that peaks within seconds and is "the worst of one's life," plus neck stiffness, vomiting, and possibly a brief loss of consciousness — is ruptured intracranial aneurysm → subarachnoid hemorrhage (SAH) until proven otherwise. Management: a non-contrast CT first (first-line, with high sensitivity within 24 hours) → if the CT is negative yet suspicion remains high, a lumbar puncture to look for xanthochromia. Bacterial meningitis has a more gradual onset with fever; migraine is recurrent and builds gradually; neither presents as thunderclap.

Vertebral osteomyelitis most commonly involves the lumbar spine (roughly 50%), followed by the thoracic spine, with the cervical spine least common; the mechanism is usually hematogenous spread, and the most common pathogen is S. aureus; MRI is the first-line imaging study, blood cultures plus biopsy culture identify the pathogen, and ESR/CRP are elevated.

At this point in the Infection and Immunity issue, the lens must flip — from "fighting the outside enemy" to "how one's own side gets attacked." Immunity really only has two jobs: recognize self, recognize non-self. Recognize too loosely, and you panic only once the outsider has already moved in — that is the story of infection told in Parts A through C. Recognize too tightly, and even your own valves, skin, glomeruli, and nerve myelin get treated as the enemy — that is the throughline of Part D. By the end of these three chapters you will see that every seemingly chaotic autoimmune disease, and every "why" behind transplantation and its drugs, strings onto this single axis of recognizing versus failing to recognize.

Once you understand sensitization, the indications and contraindications for transplantation sort themselves out naturally. Kidney transplantation is not the "last resort" for ESRD but the treatment of choice, and it is best performed before eGFR has even fallen to the point of starting dialysis — this is called pre-emptive transplantation, which yields the best long-term survival and quality of life; as for the source organ, living donation is superior to deceased donation. Contraindications must be calmly sorted into "absolute" versus "relative": the one common logic behind every absolute contraindication is that suppressing immunity would immediately worsen the patient's condition or kill them, so only active infection (such as open pulmonary tuberculosis, uncontrolled HIV, or sepsis) and active malignancy (metastatic cancer) count as absolute. Age over seventy, stably controlled HIV, and a cancer cured five years ago are all relative contraindications — the exam loves to plant these in the absolute-contraindication column as a trap.

As for whether to remove the native kidneys first, the principle is that most cases keep them — preserving whatever residual renal function remains, and sparing an extra operation. Only when the native kidney keeps causing trouble is it taken out: symptomatic polycystic kidney disease, which occupies space and keeps bleeding or getting infected; recurrent pyelonephritis or infected stones, for fear that suppressing immunity afterward will trigger sepsis; renal malignancy, which requires removing the source of cancer; and refractory proteinuria or hypertension that drugs simply cannot control. The exam loves to slip in "hypertension controllable with medication" — that is not an indication for removal; do not operate rashly.

Will the original kidney disease come back in the new kidney? This is another frequently tested angle. The one to remember most is that focal segmental glomerulosclerosis (FSGS) has a high recurrence rate and can recur as early as a few hours after surgery; dense deposit disease (DDD), IgA nephropathy, and diabetic nephropathy can all come back to bite the new kidney; lupus nephritis does not "never" recur — a minority return in the guise of a membranous lesion, so do not memorize it as "SLE is cured for good after transplant." One more item is often mistakenly answered as a sensitizing factor — eosinophilia is not an immunologic risk factor for kidney transplantation; it has nothing to do with HLA sensitization and does not affect matching or induction decisions.

Think of hypersensitivity as the four ways immunity throws a punch, and the whole plot unfolds. Type I is IgE sitting sentry on the surface of mast cells; the moment allergen cross-links it, histamine, leukotrienes, and prostaglandins burst out instantly, producing bronchospasm, vasodilation, and plasma leakage — asthma, allergic rhinitis, urticaria, and the most severe form, systemic anaphylactic shock, all belong to this type. The exam's favorite question is which type asthma belongs to: asthma is IgE-mediated Type I, not Type II. Type II is antibody sticking directly onto an "antigen on the cell surface," then recruiting complement and phagocytes to eat the cell — so in ITP the antibody grabs GPIIb/IIIa and GPIb on the platelet surface, in autoimmune hemolysis the antibody grabs red blood cells, in Goodpasture syndrome the antibody grabs the pulmonary and renal basement membrane, and in Graves' disease TRAb stimulates the TSH receptor — every one of these cell-surface stories is Type II. Type III is antigen and antibody first binding into a complex in the circulation, then depositing in vessel walls, glomeruli, skin, and joints — wherever it deposits, inflammation follows; SLE, serum sickness, post-streptococcal glomerulonephritis, and the Arthus reaction all run on this mechanism. Type IV dispenses with antibody entirely and is carried out directly by T cells and macrophages; it takes forty-eight to seventy-two hours to react, hence its other name, delayed-type — the tuberculin skin test, contact dermatitis, and transplant rejection all depend on it. In one line: Type I is IgE, Type II is cell-surface antibody, Type III is complex deposition, Type IV is T cells with no antibody at all.

Every bit of SLE's destruction runs the length of the chain above. In blood vessels you see the classic fibrinoid necrosis; in the heart you see Libman-Sacks endocarditis — small, non-bacterial, irregular vegetations growing on "both surfaces" of a valve, favoring the mitral valve above all; do not confuse it with infective endocarditis. In the kidney you see immune-complex deposition, of which Class IV diffuse proliferative is the most severe, with the classic lesion being the wire-loop lesion — do not misfile it under Class V membranous. In the skin, you see a positive lupus band test. The deposit in lupus nephritis is an immune complex, not anti-GBM — reverse this line and every drug choice that follows will be wrong too.

Complement is consumed right along with the deposits, so asking "which component gets consumed" tells you, in reverse, which pathway the lesion is taking. The classical pathway is triggered by antigen-antibody complexes, running C1 → C4 → C2; the alternative pathway is spontaneous hydrolysis on a pathogen's surface, starring C3, factor B, and properdin; the lectin pathway is triggered by MBL binding mannose; all three converge on C3 → C5 → MAC. The culprit in dense deposit disease (DDD, type II C3 glomerulopathy) is C3 nephritic factor (C3NeF), an autoantibody whose job is to "stabilize" C3 convertase rather than inhibit complement, keeping the enzyme perpetually active and burning through C3, factor B, and properdin — but because it runs only through the alternative pathway, C1 stays normal. So when a question shows "C3↓ but C1 normal," think DDD; if C1q or C4 is also low, that means the classical pathway has been activated, usually in an immune-complex disease like SLE. Master this reverse-inference route and you never need to memorize disease-complement pairings by rote.

ITP is the clean textbook example of Type II: autoantibody IgG grabs GPIIb/IIIa and GPIb on the platelet surface, splenic macrophages engulf the platelet through their Fc receptors, and the platelet count falls. The exam loves to ask about the bone marrow — megakaryocytes are actually "increased" (a compensatory response); remember it as decreased and you have fallen for the trap. It favors women of childbearing age; most respond to corticosteroids, with IVIG, anti-D, splenectomy, and TPO-receptor agonists as second-line options.

Two small but frequently tested points to fold in. The histologic triad of IgG4-related disease is lymphoplasmacytic infiltrate (IgG4+ plasma cells), storiform fibrosis, and obliterative phlebitis; eosinophils are common, but the finding you are least likely to see is neutrophils — neutrophils are the marker of acute suppurative inflammation and have no business in this chronic fibrosing disease. In myasthenia gravis, the most common thymic finding is not thymoma but lymphoid follicular hyperplasia (roughly 70–80%), with only about 15% being thymoma; the mechanism is a Type II anti-AChR antibody, the classic presentation is weakness that worsens with activity, and thymectomy can improve the disease.

The trick to skin-infection questions is to first sort the pathogen into one of four bins — virus, bacterium, fungus, parasite — then let distribution pattern, Wood's lamp fluorescence, and KOH microscopy pinpoint it for you. Blisters along a single unilateral dermatome, not crossing the midline, mean VZV reactivation: the virus normally hides in the dorsal root ganglia, and once immunity dips it travels back out along the sensory nerve — so it never crosses the midline, and the pain is especially severe at night. Giving an acyclovir-class antiviral within seventy-two hours can reduce post-herpetic neuralgia. HSV and herpes zoster differ in distribution: HSV favors the lips and genitals and does not follow a chest or abdominal dermatome; impetigo instead shows honey-colored crusting, with no blisters and no dermatomal pattern.

Wood's lamp fluorescence is a free point on the exam, but you need to understand the logic behind it: erythrasma is a bacterial infection (Corynebacterium minutissimum); the organism produces a metabolite called coproporphyrin III, which fluoresces coral-red. KOH microscopy is negative, and treatment is erythromycin or a topical antibiotic — no steroid is needed, although topical azole antifungals also work because they are active against gram-positive bacteria. Tinea versicolor is caused by the fungus Malassezia and fluoresces yellow-green or gold; some tinea capitis (Microsporum) fluoresces green; vitiligo shows enhanced white fluorescence from melanin loss and involves no infection at all. Erythrasma is therefore the condition most often mistaken for a fungal infection — remember it is bacterial, and treated with an antibiotic.

Scabies is confirmed by microscopy showing the mite, its eggs, or fecal pellets (scybala) — any one is sufficient; you need not see a live mite. The incubation period is a frequent trap: a first infection takes four to six weeks before it itches, so "itching the day after contact" absolutely cannot be diagnosed as scabies (a re-infection, already sensitized, can itch within one to three days). The favored sites in adults are the finger webs, wrists, and genitals, usually sparing the scalp; a mite can survive twenty-four to seventy-two hours off the human body. The treatment of first choice is permethrin 5%, relatively safe in pregnancy and children; lindane (γ-BHC) acts on the mite's GABA receptor and is neurotoxic, contraindicated in pregnant women, infants, and patients with epilepsy — a statement that "lindane is safe for use in pregnant women and children" is wrong. Oral ivermectin is reserved for crusted (Norwegian) scabies or large-scale outbreaks.

The most common opportunistic fungus in immunocompromised hosts is Candida, ranging from oral thrush and esophageal candidiasis all the way to invasive candidemia. For a child with a scaly facial white patch and no fluorescence under Wood's lamp, do KOH microscopy first to rule out a fungal cause — do not jump straight to biopsy or culture. One last point that is often confused: warts are caused by HPV, a keratinocyte infection, not Staphylococcus aureus.

The crux of autoantibody questions is "antibody → disease → exception." Anti-dsDNA and anti-Sm are highly specific for SLE, and dsDNA also tracks with nephritis activity; anti-SSA/Ro, though common in Sjögren syndrome, is also seen in SLE and neonatal lupus, so it is less specific — the antibody that is truly "seen almost only in Sjögren" and is the most specific is anti-SSB/La; this is the point examiners most often swap to bait a wrong answer. Anti-histone is the marker of drug-induced lupus erythematosus (DILE), positive in over 95% of cases. Anti-Scl-70 (topoisomerase I) is linked to diffuse systemic sclerosis, often with pulmonary fibrosis; anti-centromere is linked to the limited (CREST) form of sclerosis, which carries a better prognosis. ANA is a broad screening test, positive in most SLE and sclerosis but not specific.

Drug-induced lupus can be locked down with one sentence: anti-histone (+), dsDNA (−), complement normal, little renal involvement, resolves with drug withdrawal. The mnemonic for the culprit drugs is "HIP" — Hydralazine, Isoniazid, Procainamide — plus minocycline and anti-TNF agents. The sex ratio is completely different from that of SLE: primary SLE affects far more women than men, but because the population taking these drugs is more often male, DILE's sex ratio is close to even, and renal and CNS involvement are also less frequent and milder. The single most important step is stopping the drug; most cases resolve on their own, with no need for long-term immunosuppression.

The most common subtype of cutaneous lupus erythematosus is discoid lupus erythematosus (DLE), the chronic cutaneous form; only about 5% of cases progress to systemic SLE — do not memorize it as "most of them turn into SLE." The cutaneous ACR diagnostic items for SLE are these four: malar rash, photosensitivity, discoid rash, oral ulcers; urticaria is not among them, and it is a common distractor option.

The core feature of systemic sclerosis is Raynaud phenomenon (over 90%), together with skin thickening and hardening, dysphagia from esophageal smooth-muscle fibrosis, ANA positive in over 95%, and potential progression to pulmonary fibrosis and renal crisis. The heliotrope sign (a violet rash on the upper eyelids) is a feature of dermatomyositis, not sclerosis — this is a trap commonly pinned on the wrong disease. Treatment follows the mechanism accordingly: skin sclerosis (dermal fibrosis) is treated with UVA1 phototherapy (mainly for localized scleroderma/morphea; for systemic sclerosis skin, MTX or MMF first), which penetrates the dermis and suppresses fibroblasts; digital ulcers are a vascular lesion and call for a vasodilator (prostacyclin/iloprost, a calcium-channel blocker), not a topical steroid; telangiectasia has no drug that effectively reverses it, and calcinosis currently has no standard treatment.

This chapter's central thread is: which layer has split, which segment has failed, where is the source of infection. The same immune system, mistaking its target, becomes blisters in the skin, demyelination in the nerve, and an unpassable stone with infection in the kidney. Get the "layer" straight, and the questions solve themselves instantly.

The clinical presentation of the two blistering groups can be derived directly from the mechanism. Because pemphigus involves intercellular loss of attachment, its blisters are flaccid and rupture easily, erosions are extensive, and the Nikolsky sign is positive (a finger's push and the skin sloughs off); the distribution of desmoglein determines where the damage strikes first — Dsg3 is found mainly in the oral mucosa and the basal layer of the epidermis, while Dsg1 is in the upper epidermis. So in pemphigus vulgaris (PV), the antibody targets Dsg3 with or without Dsg1, the oral mucosa is affected first and becomes the presenting sign, the split occurs just above the basal layer (suprabasal), and the basal cells line the floor of the blister like tombstones. Pemphigus foliaceus (PF) targets Dsg1 only, splitting at the most superficial subcorneal/granular layer; because Dsg3 is untouched, there are no mucosal lesions, only superficial scaling. In one line: anti-Dsg3 gives you mucosal involvement plus a suprabasal split; anti-Dsg1 alone stays superficial, with no mucosal involvement. Pemphigoid, by contrast, splits beneath the epidermis; its blisters are tense and hard to rupture, the Nikolsky sign is negative, it favors the elderly, and it itches intensely.

DIF is the key to telling the two groups apart: an intercellular, chicken-wire pattern of IgG means pemphigus; a linear IgG/C3 pattern along the basement membrane means pemphigoid or EBA. See "deposit at the DEJ" and pemphigus is off the table. Bullous pemphigoid (BP) and epidermolysis bullosa acquisita (EBA) both show a subepidermal blister on H&E, with almost no way to tell them apart morphologically; you must rely on salt-split skin DIF: BP's antibody stains the roof of the split (epidermal side), because BP180/230 sit in the lamina lucida; EBA's antibody stains the floor of the split (dermal side), because type VII collagen sits in the anchoring fibrils. Clinically, BP presents as severe pruritus with tense blisters in the elderly, while EBA presents as blisters at sites of trauma, with scarring and milia.

The exam often brings in dermatomyositis as a supporting character in the differential for blistering disease. Gottron papules (violaceous papules over the dorsal finger and metacarpophalangeal joints) are the pathognomonic feature of dermatomyositis, while the heliotrope rash (violaceous edema around the eyelids) is highly suggestive but not pathognomonic — these two are frequently swapped. Dermatomyositis has two more easily tested points: it often causes noticeable pruritus (especially of the scalp — not "no pruritus" as sometimes claimed), and it is associated with malignancy (adults require a work-up).

The core to solving any CSF-analysis question is to look at the glucose ratio (CSF/serum) first. A normal ratio is above 0.6; a low ratio is nearly synonymous with bacterial or tuberculous/fungal disease. Add the dominant cell type and the protein level, and the diagnosis sorts itself: low glucose plus neutrophils plus protein over 100 = bacterial; normal glucose plus lymphocytes = viral; low glucose plus lymphocytes plus protein over 100 = tuberculous or fungal. One small calculation that is often tested: WBC 20 (predominantly lymphocytes), protein 45 (considered normal), glucose 50 with a serum glucose of 80 — a ratio of 0.63 — this is viral meningitis.

The sequencing of management in acute bacterial meningitis is the exam's favorite target. The chain of reasoning is: suspicion → blood cultures → antibiotics immediately (with or without steroids) → CSF afterward. Never delay antibiotics to wait for Gram stain or culture results — every hour of delay raises mortality. When to get a CT before the LP: focal neurologic signs, marked change in consciousness, papilledema, immunocompromise, or seizures, to guard against herniation after the LP; but even when a CT is needed, the sequence is still blood cultures and antibiotics first, then the CT. Dexamethasone must be given before or at the same time as the first dose of antibiotics (for adults with suspected pneumococcal disease) to reduce neurologic sequelae and death. Empiric therapy in adults is ceftriaxone (a third-generation cephalosporin) plus vancomycin to cover resistant pneumococcus; add ampicillin for patients over fifty or the immunocompromised to cover Listeria; neonates receive ampicillin plus cefotaxime. Note that the classic triad (fever, neck stiffness, altered mental status) is present together in fewer than 50% of cases — do not rule out the diagnosis just because not all three are present. The Cushing reflex of raised intracranial pressure is a "slow" heart rate plus high blood pressure plus irregular breathing, not a fast heart rate — this is another direction that is often remembered backward.

Specific pathogens each carry their own "fingerprint." HSV encephalitis loves to burn out hemorrhagic necrosis in the anterior temporal lobes, because HSV-1 travels retrograde along the olfactory and trigeminal nerves; MRI shows temporal-lobe hemorrhagic necrosis, and the CSF occasionally shows red cells and lymphocytes. Start acyclovir on suspicion alone; do not wait for PCR. The diagnostic combination for neurosyphilis is a positive serum treponemal test (MHA-TP/TPHA/FTA-ABS) confirming past infection, plus a positive CSF VDRL confirming CNS involvement; CSF VDRL is highly specific but poorly sensitive (a negative result cannot exclude the diagnosis), while CSF TPHA is sensitive but not specific; treatment is IV penicillin G. The classic three stages of a spinal epidural abscess are: severe back pain with fever → radicular pain → loss of motor and sensory function in the limbs; management is MRI plus emergency decompressive drainage plus antibiotics. Tuberculous meningitis should be treated with added dexamethasone (it is not contraindicated), which reduces death and sequelae; some remember it backward as "steroids are contraindicated" — that is wrong.

The key to prion disease/CJD is that PrPSc is a misfolded protein containing no nucleic acid at all; ordinary autoclaving (even at 134°C) cannot destroy it — it requires strong alkali or specialized high-pressure protocols. Classic CJD strikes at an average age of sixty-eight, with rapidly progressive dementia plus myoclonus; the EEG shows periodic sharp-wave complexes (PSWC, generalized triphasic waves at roughly 1 Hz), and CSF 14-3-3 or RT-QuIC assists the diagnosis. Variant CJD (vCJD) is linked to bovine spongiform encephalopathy ("mad cow disease") beef, favors a younger population (averaging about twenty-nine years old), and psychiatric symptoms often precede neurologic ones. The exam loves to mix up the two "periodic" EEG patterns: HSV encephalitis shows PLEDs (periodic lateralized epileptiform discharges, temporal); CJD shows PSWC (generalized triphasic waves at about 1 Hz) — neither is a feature of neurosyphilis.

Myelin is the nerve's insulating tape, letting the electrical signal conduct by leaping from node to node; damage the tape and conduction slows or even fails. Central myelin is made by the oligodendrocyte (MS, NMOSD, MOGAD, CPM, and ADEM all belong here), while peripheral myelin is made by the Schwann cell (GBS, CIDP). When a given cell type fails, the symptoms land in the territory that cell is responsible for — grasp this zoning logic and the questions stop being a jumble.

The three musketeers of central autoimmune disease — MS, NMOSD, and MOGAD — are a high-frequency differential. MS has no specific antibody (look for oligoclonal bands), its spinal-cord lesions are usually short (fewer than three segments), its optic neuritis is unilateral and milder, its female-to-male ratio is about 2:1, its incidence is higher at high latitudes (the vitamin D hypothesis), and it has few comorbidities. NMOSD's antibody is anti-AQP4 (which attacks the AQP4 water channel on the astrocyte, not myelin itself), its spinal-cord lesions are long (≥3 vertebral segments, LETM), its optic neuritis is bilateral, severe, and prone to blindness, its female-to-male ratio is about 9:1, and it frequently coexists with autoimmune diseases such as MG, Sjögren syndrome, and SLE. MOGAD's antibody is anti-MOG (which attacks MOG on the surface of myelin); its lesions are often long-segment but recover better, it relapses but carries a better prognosis, and its sex distribution is close to even. MS's dissemination in space and time (DIS + DIT) is the core of its diagnosis — lesions must appear in different locations and at different times, which is exactly where the word "multiple" comes from.

Interpreting MS's CSF involves a few frequently tested numbers: total protein is mildly elevated, usually under 50, and never over 100; say "over 100" and you are wrong. White cells are predominantly lymphocytic and fewer than 50; glucose is normal; positive oligoclonal bands (with a negative serum sample) are a feature of MS, representing intrathecal IgG synthesis. One classic trap: thymoma is associated with MG, not a comorbidity of NMOSD — it frequently shows up as a wrong answer option.

Metabolic demyelination has two commonly tested conditions. Osmotic demyelination (CPM/ODS) results from correcting chronic hyponatremia too quickly: brain cells have already adapted to the low osmolality, and a sudden rise in serum sodium causes them to lose water rapidly, dissolving the myelin of the pons. The safe rate is a rise in sodium of no more than 8–10 mEq/L per 24 hours; a question giving "a rise of 20 mEq/L in six hours" describes correction that is too fast. Symptoms usually appear one to five days after correction — not immediately, and not two weeks later. Subacute combined degeneration (SCD, from B12 deficiency), due to long-term veganism, pernicious anemia, or gastrectomy, damages the posterior columns, lateral columns, and peripheral nerves all at once, so vibration sense and proprioception are lost while pain and temperature sensation are preserved; corticospinal-tract damage produces a spastic gait plus a positive Babinski sign plus increased knee-jerk reflexes, while peripheral-nerve damage weakens the ankle reflex — this odd mixed UMN-and-LMN combination of "increased knee-jerk plus diminished ankle-jerk," together with lost vibration sense, preserved pain and temperature sensation, and a history of veganism, is the identifying key to SCD; B12 replacement is all that is needed.

Classifying urinary stones by "composition" is what brings the topic to life. Calcium oxalate is the most common (about 70%), radiopaque, and independent of urine pH; its risk factors are high urinary calcium, high urinary oxalate, and low urinary citrate. There are two counterintuitive points you must know here. First, calcium oxalate stones should not be managed by restricting dietary calcium — because calcium in the gut normally binds oxalate and carries it out together; lowering dietary calcium instead raises free oxalate in the gut, which then gets absorbed into the urine, so stone risk rises rather than falls. The right approach is to maintain normal dietary calcium, restrict sodium and animal protein, and supplement potassium citrate. Large doses of vitamin C are metabolized into oxalate and also raise risk. Second, only uric acid stones and cystine stones can be dissolved by alkalinizing the urine; calcium oxalate's solubility barely changes across the physiologic pH range, so pH adjustment is not needed.

Uric acid stones are radiolucent (invisible on X-ray), form in acidic urine (<5.5), and are linked to gout, a high-purine diet, and acidic urine. The three pillars of treatment are alkalinizing the urine (potassium citrate or sodium bicarbonate, target pH 6.5–7.0), reducing production (allopurinol or febuxostat, inhibiting xanthine oxidase), and a low-purine diet. A high-frequency trap: uricosuric drugs (probenecid, benzbromarone) are contraindicated for uric acid stones, because they push more uric acid into the urine and worsen the stone. Struvite (magnesium ammonium phosphate) stones form when urease-producing bacteria (especially Proteus) break urea down into ammonia and alkalinize the urine; the classic form is a staghorn calculus, and management is antibiotics plus complete stone removal. Cystine stones are a rare genetic disease, managed with urinary alkalinization plus tiopronin/penicillamine.

The principles of stratifying and managing UTI: asymptomatic bacteriuria (ASB) is, as a rule, not treated; using antibiotics indiscriminately only breeds more resistance. Only two situations warrant treatment: pregnant women (about a 30% risk of progression to pyelonephritis, plus increased risk of preterm birth and low birth weight), and before an invasive urologic procedure (such as TURP or stone surgery, to prevent bacteremia). ASB in the elderly, in diabetics, and in patients with an indwelling catheter does not require treatment — this is a frequent wrong-answer option. Whether the setting is community-acquired cystitis or pyelonephritis, the most common pathogen is always E. coli (about 75–85%), followed by Klebsiella and Proteus (a urease producer, linked to struvite staghorn calculi), with S. saprophyticus seen in young women.

The decision tree for ureteral stones: a stone smaller than 5–6 mm, with no infection or threat to renal function, can be observed conservatively along with increased fluid intake, with an α-blocker (tamsulosin) added to aid passage — this is called medical expulsive therapy. For a larger stone, one that is impacted, or one in the distal ureter, the first choice is URS (ureteroscopy), because its overall stone-free rate beats ESWL. ESWL is not first-line for ureteral stones; it is used only for select smaller stones in the upper ureter, and it localizes uric acid stones poorly since they are radiolucent. Reflux nephropathy arises from recurrent childhood UTIs combined with vesicoureteral reflux, and imaging shows asymmetric renal atrophy with an irregular contour; describing it as "symmetric atrophy" is wrong (symmetric atrophy is seen more with systemic causes such as chronic glomerulonephritis).

90% of all questions on immunologic drugs fall on the "suppressive" side (autoimmunity, transplantation, allergy); the remaining 10% fall on the "enhancing" side (vaccine adjuvants, echinacea, IFN, checkpoint inhibitors). Biologics are classified by "which molecule they intercept," and the name tells you the target: -mab is a monoclonal antibody, -cept is a fusion protein (a decoy receptor), and -tinib is an oral small-molecule kinase inhibitor.

String the core drugs into one mechanism map: omalizumab is an anti-IgE monoclonal antibody that binds the Fc-epsilon segment of circulating IgE, preventing IgE from engaging FcεRI on mast cells and basophils; it is used for moderate-to-severe allergic asthma, chronic spontaneous urticaria, and peanut allergy. Etanercept is a fusion protein of a TNF receptor plus Fc (a decoy receptor that neutralizes TNF); infliximab/adalimumab are anti-TNF monoclonal antibodies — both hit TNF, but by different mechanisms. Rituximab targets CD20 to clear mature B cells, used in lymphoma, RA, and ANCA vasculitis. Tocilizumab targets the IL-6 receptor, used in RA and in cytokine release syndrome (CRS) after CAR-T therapy. Secukinumab targets IL-17A and ustekinumab targets the p40 subunit shared by IL-12/23, both used for psoriasis.

Every TNF-α inhibitor requires screening for latent tuberculosis (LTBI) and hepatitis B (HBV) before starting, because both can be "reactivated" — a high-frequency point of clinical safety; TNF inhibitors can also trigger demyelination and drug-induced lupus, on top of raising the risk of opportunistic infection. The first-choice conventional DMARD for RA is still methotrexate (oral) — do not jump straight to a biologic just because you see "RA." JAK inhibitors (tofacitinib, baricitinib) are oral small molecules that block intracellular JAK-STAT signaling, an entirely different class from biologics (injected, intercepting extracellular molecules).

Differentiating the transplant-immunology drugs is also high-frequency. Cyclosporine binds cyclophilin and tacrolimus binds FKBP, but both inhibit calcineurin and thereby block IL-2 production; sirolimus works through mTOR, inhibiting signaling downstream of IL-2 rather than calcineurin, which is why it has the lowest nephrotoxicity. Their signature side effects must be kept distinct: cyclosporine causes nephrotoxicity plus gingival hyperplasia plus hirsutism plus hypertension; tacrolimus causes nephrotoxicity plus hyperglycemia/new-onset diabetes plus neurotoxicity; sirolimus causes hyperlipidemia plus myelosuppression, but low nephrotoxicity.

Questions on the immune-enhancing side are few, but echinacea is a natural immunomodulatory herb that activates macrophages, raises NK- and T-cell activity, and promotes IFN secretion; it is used as an adjunct to shorten the course of upper respiratory tract infection. It belongs to the enhancing side — do not confuse it with the suppressive drugs. IFN-α is antiviral (for hepatitis B/C) and used against certain tumors; a vaccine adjuvant (alum) enhances the response to antigen.

The core of β-lactams and carbapenems is inhibiting PBP (transpeptidase) so the cell wall cannot cross-link, and the bacterium lyses. The main cause of resistance is β-lactamase — an enzyme the bacterium secretes that hydrolyzes the β-lactam ring; so amoxicillin's resistance is mainly due to β-lactamase, solved by adding clavulanate, not by PBP mutation (PBP mutation is the mechanism of MRSA — do not confuse the two). Imipenem must be paired with cilastatin: imipenem is hydrolyzed in the renal proximal tubule by dehydropeptidase-I (DHP-I), which both lowers its activity and produces a nephrotoxic metabolite; cilastatin is a DHP-I inhibitor, combined into Primaxin to protect the drug — think of cilastatin as the bodyguard who blocks the kidney's pair of scissors so imipenem never gets cut apart.

Aminoglycosides work by binding the 30S subunit irreversibly, causing mRNA misreading. They require oxygen to enter the bacterial cell, so they are naturally resistant against anaerobes. The synergy story is elegant: a β-lactam (such as cefepime) breaks down the cell wall first → only then can an aminoglycoside (such as amikacin) get in — the two act synergistically against organisms such as Pseudomonas and other gram-negatives (enterococci are intrinsically resistant to cephalosporins, so they need ampicillin or vancomycin + gentamicin); a memory hook is "breach the wall, and the gun can get through." Their signature toxicity is nephrotoxicity plus ototoxicity (irreversible hearing loss and vestibular injury). Their dosing strategy is also distinctive: aminoglycosides are concentration-dependent killers — the higher the peak, the better the effect — and they have a post-antibiotic effect, so clinical practice uses a single large dose once daily, balancing efficacy against reduced nephrotoxicity; β-lactams are time-dependent, their efficacy determined by the "time" the concentration stays above the MIC, so they must be dosed multiple times or infused over a longer period to sustain the blood level.

Why do sulfonamides kill selectively? Because the human body cannot synthesize folate on its own and takes it directly from food, whereas bacteria must synthesize it themselves. Sulfonamides inhibit dihydropteroate synthase, and trimethoprim inhibits dihydrofolate reductase; together the two steps block de novo folate synthesis, harming only the bacterium and not the human host. Signature side effects: Stevens-Johnson syndrome, hemolysis in G6PD deficiency, and kernicterus in the newborn.

The core concept of antiviral drugs is prodrug activation. Acyclovir is a prodrug that requires viral thymidine kinase (viral TK) to phosphorylate it to the monophosphate first, after which host cell kinases continue phosphorylating it to the triphosphate before it can inhibit viral DNA polymerase. So a latent virus does not express TK, and acyclovir has no effect on latent infection; a TK mutation is the mechanism of resistance. Oseltamivir is a neuraminidase inhibitor used against influenza, and it is not used for HIV — HIV's six drug classes are NRTIs, NNRTIs (both acting on reverse transcriptase), PIs (protease), INSTIs (integrase), fusion inhibitors, and CCR5 antagonists; as of 2026, first-line regimens are mostly built on an INSTI backbone (bictegravir, dolutegravir). Tenofovir, with long-term use, has to date shown no significant resistance mutations and outperforms adefovir and lamivudine, but it is less effective against adefovir-resistant strains (not completely unaffected by them).

The four major antifungal targets must be kept distinct. Polyenes (amphotericin B, nystatin) work by binding ergosterol directly → forming pores in the membrane → electrolyte leakage kills the fungus; their signature side effects are nephrotoxicity and infusion-related rigors. Azoles (fluconazole, voriconazole) work by inhibiting 14-alpha-demethylase (a CYP enzyme) to block ergosterol synthesis — attacking the same cell membrane, but one binds the finished product while the other cuts off the upstream supply; their side effects are extensive drug interactions and hepatotoxicity. Echinocandins (caspofungin) inhibit synthesis of β-1,3-glucan in the cell wall and carry a high safety margin. Flucytosine (5-FC) interferes with fungal DNA/RNA synthesis and is often combined with amphotericin to treat cryptococcal meningitis.

The core to solving pleural-effusion questions is Light's criteria — meeting any one of them classifies the fluid as an exudate: fluid-to-serum protein ratio > 0.5, LDH ratio > 0.6, or effusion LDH > two-thirds of the upper limit of normal for serum. The logic behind it: a transudate is a pressure problem (heart failure, cirrhosis, nephrotic syndrome) squeezing fluid out, with low protein; an exudate results from inflammation, tumor, or blocked lymphatic drainage raising vascular permeability, so both protein and LDH are high. Both malignant and infectious effusions are exudates.

The causes of malignant pleural effusion must be sorted by sex: the single most common overall is lung adenocarcinoma (~35–40%); the most common in women is breast cancer; other common causes include lymphoma and mesothelioma (asbestos-related); melanoma is seen occasionally but is not common. Management principles: drainage relieves dyspnea but does not prolong survival; for recurrent effusion, the first choice is talc pleurodesis, which is the most effective option (an indwelling pleural catheter is an equally acceptable first-line option when the lung is expandable) — a point commonly disguised as "the least effective" in trap answers. A malignant pleural effusion in lung cancer, confirmed by cytology or biopsy, is staged M1a (Stage IV, not T4) and is not curable by surgery; but a "benign/paraneoplastic" effusion (cytology negative) does not affect surgical staging.

The high-frequency decision point for complicated parapneumonic effusion or empyema: if an exudative effusion accompanying pneumonia progresses to pH < 7.2, low glucose, markedly elevated LDH, or frank pus or a positive culture on aspiration, it is complicated, and antibiotics alone are not enough — tube thoracostomy is required; a loculated effusion may need an intrapleural fibrinolytic added, or VATS debridement. Lung abscess most commonly arises via aspiration (anaerobes, in a patient with altered consciousness or dysphagia), but it can also spread hematogenously — Staphylococcus aureus bacteremia can spread to the lung hematogenously and produce multiple abscesses, so the statement "lung abscess has nothing to do with staphylococcal bacteremia" is false.

Aspergilloma (fungus ball) forms when Aspergillus colonizes an existing cavity (often an old tuberculous cavity) into a ball, presenting mainly with hemoptysis, which can be fatal. The key decision: blood flow inside the cavity is poor and antifungal drugs cannot penetrate it, so systemic antifungal therapy has limited effect; symptomatic (hemoptysis) cases are best treated with surgical resection rather than long-term antifungal drugs. Remember one line: "a ball stuck in a hole" is a structural problem — it calls for surgery, not medication.

Three congenital pulmonary cystic lesions must be remembered. A bronchogenic cyst is recommended for resection even when asymptomatic, for fear of infection, compression, hemorrhage, and malignant transformation. CPAM/CCAM carries malignant potential (able to progress to pleuropulmonary blastoma), and resection is recommended for most cases. Pulmonary sequestration has no connection to the normal airway and receives systemic arterial blood supply (often from the aorta); recurrent infection calls for surgery, and the blood supply must be identified beforehand to avoid massive hemorrhage. The core concept: these congenital lesions do not resolve on their own, and the risk of recurrent infection and malignant transformation means resection is favored even when asymptomatic; "CPAM has no malignant potential" and "an asymptomatic bronchogenic cyst needs no treatment" are both wrong answer choices.

Apical lung tumors and mediastinal syndromes can be strung together in one line: Pancoast syndrome (an apical tumor invading the brachial plexus) produces shoulder and arm pain with hand weakness; Horner syndrome (Pancoast invading further into the cervical sympathetic chain) produces ipsilateral ptosis plus miosis plus anhidrosis; SVC syndrome (a superior mediastinal tumor compressing the superior vena cava) produces facial and upper-limb swelling with distended neck veins. Pseudopolycythemia — dehydration concentrating the plasma — has nothing to do with thoracic pathology, and it is often the correct answer to "which is least related"; it is true hypoxia-driven secondary polycythemia that is actually associated with chronic lung disease. Sarcoidosis is a systemic disease of non-caseating granulomas, with over 90% showing lung involvement (bilateral hilar lymphadenopathy plus intrapulmonary nodules); the contrast with tuberculosis's caseating granulomas is a must-know point — do not reverse it.

The differential for bone tumors can be settled for the most part by "age + location + imaging pattern." The highest-frequency tumor, osteosarcoma, is malignant, occurring in adolescents aged 10–20, at the metaphysis of a long bone around the knee, with nighttime pain, and an X-ray showing a Codman triangle and a sunburst pattern; ALP is elevated, reflecting osteoblastic activity. About 15–20% already have hematogenous lung metastasis at diagnosis (the lung is the first site of metastasis), so chest imaging is mandatory at initial workup, and treatment is preoperative chemotherapy plus wide excision plus postoperative chemotherapy. Ewing sarcoma also favors ages 5–15, occurring at the diaphysis of a long bone and the pelvis, with an onion-skin periosteal reaction and t(11;22). Chondrosarcoma occurs in patients over 40, in the axial skeleton/pelvis/proximal femur, with cartilaginous-matrix calcification in a popcorn pattern; grading requires all three of clinical findings (degree of pain), radiology (size, extent of invasion), and histopathology together — none can be skipped, because low-grade chondrosarcoma is difficult to distinguish from benign enchondroma by pathology alone.

The benign tumors must also be known. Osteoid osteoma occurs in adolescents, at the diaphysis of a long bone, with nighttime pain markedly relieved by NSAIDs and a radiolucent nidus — this NSAID relief is the identifying key. Osteoblastoma resembles osteoid osteoma but is >2 cm, favors the posterior elements of the spine, and responds poorly to NSAIDs, requiring surgery. Osteochondroma is the most common benign bone tumor, occurring in adolescents at the metaphysis of a long bone, presenting as a bony projection capped with cartilage; its overall malignant-transformation rate is under 1%, but a lesion on a "flat bone" (pelvis, scapula) carries a higher risk of malignant transformation (→ chondrosarcoma) than one on a long tubular bone of the limbs; multiple hereditary exostoses (MHE) is autosomal dominant with high penetrance, produces multiple lesions, and carries a higher risk of malignant transformation. Giant cell tumor occurs at ages 20–40, at the epiphysis of a long bone (this location is critical — keep the three separate: osteosarcoma's metaphysis, osteoid osteoma's diaphysis, and this epiphysis), with a soap-bubble appearance on imaging.

But there is one major premise that is often overlooked — the most common malignant bone lesion in an adult is not a primary bone tumor but "metastatic cancer." The primary cancers that most often metastasize to bone are breast, lung, thyroid, kidney, and prostate, mostly lytic; only prostate cancer, and some breast cancers, are blastic (osteoblastic/sclerotic). So for new-onset bone pain or a bone lesion in a patient over 40, think first of metastasis and multiple myeloma — do not jump straight to a primary bone tumor.

Metabolic and structural bone diseases are often confused with tumors and must be told apart. The mechanism of Paget's disease of bone is excessive bone resorption driving a compensatory increase in abnormal bone formation (both are elevated), so the pathology shows coarse, disorganized trabeculae, irregular woven bone, and cement lines in a mosaic pattern; laboratory tests show markedly elevated ALP and elevated urinary pyridinoline (a marker of bone resorption); treatment is a bisphosphonate (to suppress bone resorption) — PTH is used for osteoporosis, not for Paget's disease; do not prescribe the wrong drug. Fibrous dysplasia is normal bone replaced by fibrous tissue, appearing as ground-glass on imaging; serum calcium, phosphate, and ALP are usually normal — this is a frequently tested point, and assuming they will be abnormal is wrong; McCune-Albright syndrome is multiple fibrous dysplasia plus café-au-lait spots plus endocrine abnormalities.

One last commonly confused point: bone pain after radiation therapy = an insufficiency fracture. The scenario is low-back or sacral pain appearing after pelvic radiotherapy (following uterine/ovarian surgery), with a bone scan showing increased uptake in the sacrum or pelvis, often in the shape of an "H," the Honda sign. The mechanism is that radiation weakens the bone, and a fatigue fracture occurs under ordinary weight-bearing. The key to the differential: this is not bone metastasis, not tumor recurrence, and not degenerative spine disease — it is the finding most often misjudged as one of those.

The previous twelve chapters covered pathogen biology and immunology that apply worldwide. But Taiwan's licensing exam has another half: the epidemiology and institutions specific to this island. The same febrile patient should bring Lyme disease to mind in Boston, but dengue fever in Tainan; the same laborer with jaundice after wading through floodwater is a rare curiosity in London, but a routine differential after typhoon season in Taiwan. This chapter first sets up the skeleton of the system — the classification of notifiable communicable diseases, reporting deadlines, and a physician's statutory duty — and then walks through this island's major local diseases along the lines of its geography and seasons.

(Classification and deadlines verified as of July 2026, per Article 39 of the Communicable Disease Control Act and current announcements of the Centers for Disease Control, Ministry of Health and Welfare.)

There is something here more important than the list itself: reporting is a physician's statutory duty, not a matter of discretion. The Communicable Disease Control Act requires that when a physician, through diagnosis, treatment, or laboratory testing, discovers a notifiable communicable disease or a suspected case, they must report it to the local competent authority within the prescribed deadline; failing to report as required, or concealing a case, is subject to a fine under that same Act's penalty provisions. The deadline is counted in hours, not business days — the middle of the night, weekends, and holidays all count. So the correct clinical reflex is: report on suspicion alone; do not wait for culture confirmation — by the time the culture grows out, the transmission chain has already advanced three generations.

Dengue fever in Taiwan is a story of geography. Aedes aegypti is found south of Budai in Chiayi County, including Tainan, Kaohsiung, Pingtung, Taitung City, and Magong on Penghu; Aedes albopictus, by contrast, is distributed across lowland areas of the entire island and mountain areas below 1,500 meters (verified as of July 2026). The difference between the two lies not in their names but in their habits, and those habits directly determine transmission efficiency: Aedes aegypti prefers human blood, favors resting indoors (roughly 70–80% of the time), and bites multiple people before laying a single batch of eggs — meaning a single mosquito can string together an entire household; Aedes albopictus prefers the outdoors, usually takes a single blood meal per person, and transmits the virus less efficiently. So the answer to "why does dengue burn so fiercely in the south" is not that the south is hotter — it is that the south has a mosquito that lives in your living room, specializes in biting people, and bites several people in a single meal.

The rhythm of an outbreak can also be derived logically: Aedes mosquitoes breed in man-made containers of standing water (flowerpot saucers, discarded tires, water tanks, sagging tarpaulins), so the core of prevention is always eliminating breeding sites, not spraying insecticide; spraying kills only adult mosquitoes and cannot outlast a single generation. In 2015, Tainan saw the most severe locally acquired outbreak on record, with the nation's locally acquired cases reaching over 43,000 (verified as of July 2026) and deaths exceeding two hundred (⚠️ pending verification); in 2023, Tainan saw another large outbreak, the most severe since 2015 (verified as of July 2026).

The single most important thing to remember clinically is the three-phase timeline, and the danger period falls, of all times, exactly when the patient "feels better."

The warning signs must be memorized cold, because they decide whether the patient is hospitalized: severe abdominal pain or tenderness, persistent vomiting, clinically evident fluid accumulation (ascites or pleural effusion), mucosal bleeding, lethargy or restlessness, hepatomegaly over 2 cm, and a rising hematocrit combined with a sharp drop in platelets. Severe dengue is then defined by any one of three criteria: severe plasma leakage leading to shock (dengue shock syndrome), severe hemorrhage, or severe organ impairment (liver, heart, central nervous system).

Diagnosis splits into two stages, with the dividing line at day 5 of illness: within 5 days, test the NS1 antigen rapid test or RT-PCR (the viremic period); after day 5, test IgM/IgG (once antibodies have risen). So "a negative IgM on day 2 of illness rules out dengue" is wrong. There is no specific antiviral treatment; the mainstay is fluid management titrated to hematocrit and urine output; for fever and pain, only acetaminophen may be used — aspirin and nonsteroidal anti-inflammatory drugs (NSAIDs) are contraindicated — the platelet count is already falling, and further suppressing platelet function and the gastric mucosa invites major gastrointestinal hemorrhage. Taiwan has not yet incorporated a dengue vaccine into routine immunization (⚠️ pending verification), so prevention still relies entirely on vector-mosquito management and early reporting.

Clinically, the vast majority of infections are asymptomatic or produce only mild fever and headache, with only a very small minority progressing to encephalitis; once encephalitis develops, it presents with high fever, altered consciousness, neck stiffness, seizures, and the quite distinctive extrapyramidal syndrome (a mask-like face, cogwheel rigidity, tremor), carrying a high mortality rate and a high rate of neurologic sequelae among survivors. Diagnosis relies on IgM in serum or cerebrospinal fluid. There is no specific antiviral drug, so the entire line of defense rests on the vaccine.

In Taiwan, starting May 22, 2017, the routine childhood immunization switched from a mouse-brain-derived inactivated vaccine to a cell-culture-derived live attenuated chimeric vaccine — built on the 17D strain backbone of the yellow fever vaccine, into which the prM and E proteins of the Japanese encephalitis SA14-14-2 strain are inserted using recombinant technology; the schedule was changed to a first dose at 15 months of age and a second dose 12 months later, two doses total (verified as of July 2026). Because it is a live attenuated vaccine, the iron rule from Chapters 2 and 5 still applies: contraindicated in pregnant women and those with severe immunocompromise.

Scrub typhus is caused by *Orientia tsutsugamushi*, transmitted by chigger mites, and — echoing that favorite detail from Chapter 4 — only the larval stage of the mite bites and feeds on tissue fluid. Its distribution in Taiwan is highly regional: the incidence is highest in the east (Hualien, Taitung) and the outlying islands (Penghu, Kinmen, Matsu). Its hallmark is the eschar: a painless, itch-free ulcer with a black central crust ringed by erythema, which of all places favors the armpits, groin, waistline, and popliteal fossae — sites hidden under clothing — so a patient's claim of "I wasn't bitten" cannot rule the disease out at all; you must lift the patient's clothing and search yourself. Add fever, headache, lymphadenopathy, and a rash, and you have the classic combination. The treatment of first choice is doxycycline, with an extremely rapid response — the fever typically breaks within 48 hours; mortality is not low if the disease is inadequately treated.

Leptospirosis has already appeared in Chapter 8's discussion of travel-related infection; here we add the Taiwan perspective: it is a seasonal disease that follows typhoons and floods, entering through skin wounds or mucous membranes from standing water contaminated with rat urine, with the highest-risk groups being farmers and fishermen, sanitation workers, and volunteers or residents clearing post-flood silt. The clinical course is biphasic: fever, severe calf muscle pain, and conjunctival suffusion without discharge; the severe form, Weil's disease, is jaundice plus acute kidney injury plus a bleeding tendency. Treatment is penicillin or doxycycline.

Melioidosis is caused by *Burkholderia pseudomallei*, an organism that lives in soil and surface water; heavy typhoon rains churn the mud into an aerosol, and people become infected by inhaling it or through wound contact, which is why Taiwan's melioidosis cases always surface roughly two weeks after a typhoon. Diabetes is the single most important host risk factor (followed by chronic kidney disease, alcoholism, and chronic lung disease). On Gram stain, the classic clue is bipolar staining resembling a safety pin; treatment comes in two stages — an intensive phase with ceftazidime, or meropenem for severe disease, for at least 10 to 14 days, followed by an eradication phase of oral TMP-SMX for 3 to 6 months to prevent relapse.

Q fever is caused by *Coxiella burnetii*, transmitted via an aerosol from the placenta and amniotic fluid at the time cattle or sheep give birth, with occupational exposure (slaughtering, animal husbandry, veterinary work) being the key risk factor; reported cases in Taiwan are concentrated more in the south (the Kaohsiung-Pingtung area) (⚠️ pending verification). It is also, as covered in Chapter 8, one of the culprits behind culture-negative endocarditis, and treatment is likewise doxycycline. Hantavirus syndrome, by contrast, spreads through inhaling an aerosol formed from dried rodent excreta, with cleaning out a long-unused warehouse or basement being the classic scenario; in Taiwan the predominant form is Hantavirus Hemorrhagic Fever with Renal Syndrome (HFRS), presenting with fever, a bleeding tendency, and acute kidney injury, and it is a Category 2 disease, reported within 24 hours.

Taiwan had long been regarded as free of rabies (⚠️ pending verification) — decades had passed without a locally acquired human case or an outbreak in dogs and cats — until rabies virus was detected for the first time in a dead wild ferret-badger in July 2013, and the finding was reported to the World Organisation for Animal Health (WOAH) (verified as of July 2026). The significance of this event lies not in how many cases were involved but in the fact that it overturned the premise that "Taiwan has no rabies": from that point on, any wound from a bite or scratch by a wild carnivore (ferret-badger, masked palm civet, house shrew) must be evaluated as a possible rabies exposure, not simply treated as an ordinary injury. Policy responses that followed included free vaccination for dogs and cats in high-risk areas, expanded wildlife surveillance, and a comprehensive overhaul of the post-exposure prophylaxis protocol.

The key fact about rabies is always this: "once symptoms appear, the fatality rate is nearly one hundred percent, so everything hinges on what is done before onset." The first step of post-exposure prophylaxis (PEP) is not an injection but wound care: immediate washing with soap and copious water for at least 15 minutes, followed by disinfection with povidone-iodine — this single step alone can dramatically lower infection risk, because the virus must linger locally at the wound for a period of time before it enters a nerve. Next comes a series of rabies vaccine doses, graded to the exposure category; anyone not previously vaccinated additionally needs human rabies immunoglobulin (HRIG) infiltrated around the wound to provide immediate passive immunity (exact dosing and schedule follow the current guidelines of the Centers for Disease Control, Ministry of Health and Welfare; ⚠️ pending verification). There is one concept that must be remembered here: rabies allows no "wait and see" option, because once symptoms appear there is no cure; at the same time, the wound should not be closed immediately by suturing, to avoid sewing the virus into deep tissue.

In 1998, Taiwan suffered a major outbreak of enterovirus 71, with millions of children across the country contracting hand-foot-mouth disease or herpangina; among them, 405 cases developed severe complications and 78 children died, a case-fatality rate of 19.3% among the severe cases (verified as of July 2026). This outbreak directly reshaped Taiwan's public-health organization — the Centers for Disease Control (today the Centers for Disease Control, Ministry of Health and Welfare) was established in 1999 (verified as of July 2026). So enterovirus 71 in Taiwan is not merely a pathogen; it is an institutional turning point.

There is one detail on the prevention side that is both elegant in its mechanism and a perennial favorite on the exam: enteroviruses belong to Picornaviridae and are naked (non-enveloped) viruses — returning to Chapter 6's "PPPA is naked" logic, alcohol kills mainly by dissolving a lipid envelope, so it works poorly against an unenveloped virus like enterovirus. So in caring for a child with enterovirus disease, alcohol-based hand rub is not enough — hands must be scrubbed with soap and running water, and the environment must be disinfected with chlorine-based bleach. This is the same physical logic, in a different guise, as Chapter 9's point that "alcohol is ineffective after caring for a patient with *Clostridioides difficile*, and soap-and-water washing is required": what alcohol cannot kill — spores, naked viruses — can only be removed by physical scrubbing or an oxidizing agent. As for reporting, enterovirus infection with severe complications is a Category 3 disease, reported within one week.

Both bacillary dysentery (Shigella) and amoebic dysentery (Entamoeba histolytica) are Category 2 diseases in Taiwan, reported within 24 hours, because both travel the fecal-oral route and both can produce clustered outbreaks within institutions (residential care facilities, kindergartens, long-term care facilities). The differential between the two has already been laid out in Chapters 3 and 6; here we add only the key difference from a Taiwan perspective: Shigella's infectious dose is extremely low (as few as ten to a hundred organisms can cause disease), so a single caregiver who fails to wash their hands can be enough to bring down an entire facility — precisely why it is classified for 24-hour reporting; amoeba, by contrast, spreads via cysts in contaminated food or water, with a long incubation period and the possibility of prolonged asymptomatic cyst carriage, so a cluster has often been smoldering for months by the time it is discovered. Clinically, bacillary dysentery presents as high fever plus tenesmus plus bloody, mucoid stool, while amoebic dysentery presents as "anchovy paste" bloody stool, flask-shaped ulcers, and possible right-lobe liver abscess; treatment for the former follows susceptibility testing, while the latter requires metronidazole plus paromomycin to clear intraluminal cysts. Once either is confirmed, it also triggers the suspension and follow-up testing of workers in specific occupations such as food service, childcare, and healthcare (⚠️ pending verification; per the current disease-control operations manual).

Invasive pneumococcal disease, meanwhile, is a Category 4 disease, and its very definition is a test point: it counts only when Streptococcus pneumoniae is isolated from a sterile site (blood, cerebrospinal fluid, pleural fluid, joint fluid) — a positive sputum culture from ordinary pneumonia does not count as "invasive." The high-risk groups are children under 2, adults over 65, those with splenic dysfunction or asplenia, and the immunocompromised, and the "spleen" item echoes Chapter 1 exactly: capsular polysaccharide is a T-independent antigen, cleared mainly by marginal-zone B cells of the spleen together with opsonins, so once the spleen is gone, an encapsulated organism advances unopposed. Taiwan has already incorporated the conjugate pneumococcal vaccine into routine, publicly funded childhood immunization, and also offers publicly funded vaccination to high-risk older adults (⚠️ pending verification; the funded items and brands for each age group are adjusted according to the current immunization schedule).

Taiwan's medical licensing exam loves these three diseases, and not just because they are common — each one represents a different model of public health intervention: tuberculosis tests "how to make sure the patient actually finishes the drugs," HIV tests "how to make the people who need it most willing to walk through the door," and hepatitis tests "how one vaccine and one course of treatment can rewrite an entire generation's cancer risk." Once you have the causal logic of these three models fully worked out, the drug names and numbers will stick on their own.

Taiwan's record is worth writing into the answer: the tuberculosis incidence rate fell from about 73 cases per 100,000 population in 2005 to about 25 per 100,000 in 2025, with new cases dropping below six thousand for the first time (5,742 cases) — a cumulative decline of roughly 66% (verified as of July 2026); the goal is to move toward eliminating tuberculosis by 2035, in step with the global target. For notification purposes, ordinary tuberculosis is a Category 3 notifiable disease, reported within one week; multidrug-resistant tuberculosis is Category 2, reported within 24 hours — that gap is itself a testable point, because the public health threat level of MDR-TB sits in an entirely different class.

The second front is latent tuberculosis infection (LTBI). Pin the concept down first: latent infection is not disease, is not contagious, comes with a normal chest X-ray and a negative sputum culture — it only means the body harbors live tuberculosis bacteria held in check by the immune system, with roughly a 10% lifetime risk of progressing to active disease (substantially higher in the immunocompromised). So the point of treating LTBI is not to cure a disease — it is to defuse the fuse before it ever becomes a source of transmission. Diagnosis relies on the tuberculin skin test (TST) or interferon-gamma release assay (IGRA) covered in Chapter 6 — remember that IGRA's ESAT-6 and CFP-10 antigens come from the RD1 gene region, which has been deleted from the BCG vaccine strain, so a person who has received BCG can still test IGRA-negative; in a country like Taiwan with near-universal BCG vaccination, IGRA's edge in specificity is especially decisive.

Taiwan's current LTBI regimens have expanded from a single nine-month course of isoniazid into a menu of options: 3HP (once-weekly isoniazid plus rifapentine, 12 doses over 12 weeks), 3HR, 4R, 9H, plus the newly added 1HP (daily isoniazid plus rifapentine for one month) and 6H (verified as of July 2026). The trend is clear — shortening the course is exactly how you raise the completion rate — and the twelve doses of 3HP are still, in principle, directly observed, extending the spirit of DOTS from active tuberculosis all the way to latent infection.

Taiwan's system has three layers of tools. The first is anonymous screening and diverse testing channels (hospitals and clinics, civic organizations, self-testing kits), lowering the barrier to walking in. The second is pre-exposure prophylaxis (PrEP): a person who is not yet infected but remains at persistently high risk takes antiviral drugs on a regular schedule beforehand, so that when the virus arrives it cannot establish infection. Taiwan has run a publicly funded PrEP program since 2018; the current publicly funded groups are two — (1) the spouse or sexual partner of a person living with HIV (women prioritized), a citizen or the foreign-national spouse of a citizen living with HIV, who tests HIV-negative and is assessed as high risk; (2) citizens aged 35 or younger who test HIV-negative and are assessed as high risk — and those who use drugs to enhance sex (chemsex) and have enrolled in addiction treatment, as well as sex workers, are exempt from the age limit (verified as of July 2026). The iron rule of PrEP is that HIV-negative status must be confirmed before prescribing — giving two-drug PrEP by mistake during the window period of acute infection amounts to suppressing an infection that already exists with an incomplete drug combination, which is the textbook recipe for inducing drug-resistant mutants. The third layer is post-exposure prophylaxis (PEP), echoing the needlestick management from Chapter 9: the sooner the better, no later than 72 hours, a 4-week course.

Taiwan was once one of the world's most notorious hepatitis B high-prevalence regions, with an adult carrier rate that at one point approached 20%, and the dominant source of carriage was mother-to-child vertical transmission — a fact that dictated the entire intervention strategy. In July 1984, Taiwan became the first country in the world to vaccinate newborns of hepatitis B carrier mothers against hepatitis B; in July 1986 the program expanded to all newborns, making Taiwan the first country in the world to run a universal newborn hepatitis B vaccination policy (verified as of July 2026).

The result is textbook-grade: the pediatric hepatitis B carrier rate fell from about 10% to under 1% (the most recent figure for the vaccine generation is about 0.64%), and Taiwanese data published in 1997 in the *New England Journal of Medicine* showed a marked drop in childhood hepatocellular carcinoma incidence (verified as of July 2026) — the first proof in human history that a vaccine can prevent cancer, a full generation ahead of the HPV vaccine. The World Health Organization later drew on Taiwan's experience in shaping its global policy of universal infant vaccination.

Hepatitis C represents a different kind of victory — not through a vaccine (there is still no hepatitis C vaccine, because its E1/E2 envelope proteins are highly variable and evade neutralizing antibodies), but through drugs paired with a reimbursement policy. Direct-acting antivirals (DAAs) target three points on HCV — the NS3/4A protease, NS5A, and the NS5B polymerase — with an 8-to-12-week course, a cure rate near 99%, and minimal side effects, completely replacing the old era of interferon plus ribavirin with its brutal side effects and limited efficacy. Taiwan's key decisions came in two steps: DAAs were added to National Health Insurance coverage in 2017; starting January 2019, the requirement of "documented liver fibrosis" for reimbursement was removed, so every patient with chronic hepatitis C could be treated (verified as of July 2026). At the policy level, there is the "National Hepatitis C Elimination Policy Framework" and the 2025 Hepatitis C Elimination Program, and in 2025 Taiwan submitted an elimination report to the WHO Western Pacific Region, reaching the WHO's 2030 viral hepatitis elimination target five years ahead of schedule (verified as of July 2026). On the screening side, the once-in-a-lifetime adult hepatitis B and C screening program has been expanded (ages 45 to 79, ages 40 to 79 for Indigenous people, subject to continuing policy adjustment; ⚠️ pending verification).

One sentence captures the strategic difference between the two hepatitis viruses: hepatitis B's answer lies in the moment of birth (vaccine plus blocking mother-to-child transmission), hepatitis C's answer lies in finding the person (screening plus universal curative reimbursement). Hepatitis D, meanwhile, rides on the back of hepatitis B — it is a defective virus that borrows HBsAg for its envelope — so vaccinating against hepatitis B simultaneously prevents hepatitis D, a direct public health dividend in Taiwan from that same iron rule laid out in Chapter 6.

The previous fourteen chapters dealt with pathogens and hosts. This final chapter deals with the hospital itself — an artificial ecosystem with some of the highest antibiotic concentrations in the world, the weakest host immunity, and catheters threaded through nearly everything. Taiwan's licensing exam loves this territory because it tests knowledge and institutions at the same time: the mechanisms of resistance, the operations of infection control, and the regulations and workflows left behind by two pandemics.

Taiwan's most representative resistant organism is CRAB. According to data from the Taiwan Nosocomial Infections Surveillance (TNIS) system, the carbapenem resistance rate among *Acinetobacter baumannii* isolates in medical center ICUs rose from 59.6% in 2008 to 70.6% in the third quarter of 2017; in regional hospital ICUs it rose from 63.1% to 71.6% (verified as of July 2026). Exactly why this organism is so difficult to treat can be worked out entirely from mechanism: its resistance is usually a stack of multiple mechanisms — a class D β-lactamase (an OXA-type carbapenemase, echoing the Ambler classification from Chapter 6) + loss of outer membrane porins that keeps the drug from getting in + efflux pumps that push the drug back out — three lines of defense layered together, which is why CRAB is usually resistant to most other classes of antibiotics as well. Worse still is its ecological trait: *Acinetobacter baumannii* is extremely tolerant of desiccation and can survive for weeks on dry environmental surfaces such as bed rails, keyboards, and ventilator control panels — a fact that directly sets the priority for control: against CRAB, environmental cleaning and contact precautions matter just as much as any antibiotic.

Methicillin-resistant *Staphylococcus aureus* (MRSA) once made Taiwan one of the countries with the highest prevalence in Asia, with MRSA at one point accounting for more than half of hospital *S. aureus* isolates; in recent years the trend has been downward thanks to infection control and antimicrobial stewardship efforts (⚠️ pending verification: exact percentages depend on that year's TNIS report). The mechanism is still that same iron rule from Chapter 6: mecA encodes PBP2a — a change in the target, not enzymatic breakdown, so adding a β-lactamase inhibitor does nothing at all. Clinically, two sources must be distinguished: healthcare-associated MRSA (HA-MRSA) is seen mostly in hospitalized patients, long-term care residents, dialysis patients, and those with indwelling catheters, and carries a broad resistance profile; community-associated MRSA (CA-MRSA) often carries the Panton-Valentine leukocidin (PVL) toxin and SCC*mec* type IV, typically presenting as recurrent skin abscesses and necrotizing pneumonia in otherwise healthy young people, and usually remains susceptible to non-β-lactams such as TMP-SMX and clindamycin. The other group to watch is carbapenem-resistant Enterobacterales (CRE), with mechanisms including the class A enzyme KPC and the class B metallo-β-lactamases NDM, IMP, and VIM — class B metallo-enzymes depend on zinc ions, can hydrolyze carbapenems, and are unaffected by clavulanate — a high-frequency trap planted back in Chapter 6.

An antimicrobial stewardship program (ASP) is exactly this idea of "reducing selective pressure" turned into policy. Its core is not "use less antibiotic" but use it right — the four D's: right Drug, right Dose (correct amount and route), right De-escalation (step down the moment culture results come back), right Duration (stop when it should stop). In practice, the two main strategies are prospective audit and feedback — an infectious disease specialist or clinical pharmacist reviews cases individually and gives recommendations — and preauthorization — certain last-line antibiotics require review or infectious disease approval before they can be prescribed. Taiwan adds one more distinctive lever: National Health Insurance sets usage criteria and review mechanisms for antibiotics (especially last-line and broad-spectrum agents), and prescriptions that do not meet the approved indication may go unreimbursed — effectively using the payment system to add a brake to the ASP (⚠️ pending verification: the reimbursement rules for individual agents follow the current National Health Insurance review guidelines).

Taiwan's surveillance of healthcare-associated infection (HAI) centers on the same four categories covered in Chapter 9 — ventilator-associated pneumonia (VAP), catheter-associated urinary tract infection (CAUTI), central line-associated bloodstream infection (CLABSI), and surgical site infection (SSI). Here we fill in the definitions and the bundle care elements, because the licensing exam loves asking "which item is not part of the bundle."

Hand hygiene is the single highest-yield, lowest-cost measure among all infection control practices, and the World Health Organization's "My Five Moments for Hand Hygiene" actually falls into two clear causal groups: ① before touching a patient, ② before a clean or aseptic procedure — these two protect the patient, keeping you from carrying outside organisms onto them or into a sterile site; ③ after exposure to body fluid risk, ④ after touching a patient, ⑤ after touching the patient's surroundings — these three protect you and the next patient, keeping the organisms from this bed from traveling to another. So the order is not something to memorize — it follows logically from "protect them on the way in, protect everyone on the way out." Add one more exception from Chapter 9: when hands are visibly soiled, and after caring for a patient with *Clostridioides difficile*, alcohol-based hand rub is ineffective, and hands must be washed with soap and running water — the same physical logic also applies to non-enveloped viruses such as enterovirus.

Isolation precautions, meanwhile, are built on a foundation of standard precautions, with transmission-based precautions layered on top. The spirit of standard precautions is "treat every patient's blood and body fluids as potentially infectious," regardless of diagnosis. The distinctions and traps among the three transmission-based precautions are as follows:

Severe acute respiratory syndrome (SARS) devastated Taiwan's healthcare system in 2003, and the biggest lesson it left behind was not virology but institutional design. Three failures at the mechanistic level deserve to be written into the answer: first, the collapse of in-hospital infection control was the main engine of the outbreak's spread — with no fever screening stations, no cohorting by zone, and crossing patient-flow paths, the hospital turned from a place of treatment into an amplifier; second, the blanket lockdown confined the infected and the uninfected within the same space, violating the most basic principle of isolation — the purpose of isolation is to separate people, not to gather them together; third, a gap in reporting and information flow delayed decision-making. The institutional response that followed included a major revision of the Communicable Disease Control Act (adding the establishment and authority of the Central Epidemic Command Center, procedures for isolation and compulsory measures, and compensation for those isolated, among other provisions) (⚠️ pending verification: the exact year and article numbers of the revision), along with the creation of a nationwide network of negative-pressure isolation rooms, a hospital infection control audit system, and fever screening and patient-flow segregation mechanisms, as well as the TOCC intake routine still used in outpatient and emergency departments today — Travel history, Occupation, Contact history, and Cluster. The temperature checks, mask requirements, and segregated patient flow you see at the entrance of every Taiwanese hospital today are all legacies of that spring in 2003.

Coronavirus disease 2019 (COVID-19) was the second major operation. In the early phase of the pandemic, Taiwan's core measures were border control, home quarantine and isolation, TOCC prompts linked through National Health Insurance and immigration data, and mask rationing; among these, the mask name-based rationing system was the most institutionally distinctive element — when civilian demand suddenly spiked and the market failed, the government requisitioned production capacity and, through quota sales at pharmacies and health centers using the National Health Insurance card, turned masks from "whoever pays the most gets them" into "everyone gets a share," while still keeping priority supply for the healthcare sector. The pandemic also accumulated a great deal of public debate, including the timeline and channels for vaccine procurement, the review standards for domestic vaccines' emergency use authorization (EUA), and the role of independent procurement by civic groups and local governments (⚠️ pending verification: the exact timeline and administrative actions in each controversy) — these controversies are themselves exam material for public health ethics and governance: how efficiency, transparency, and procedural justice should be weighed against each other under emergency conditions. Worth noting too is how the institutional response wound down: as the pandemic eased, COVID-19 was reclassified from a Category 5 to a Category 4 notifiable disease, reported under the name "COVID-19 with severe complications" (verified as of July 2026), and the Central Epidemic Command Center has since been disbanded.